Fan height measuring method and device based on high-resolution satellite image shadow

Abstract

The invention discloses a fan height measuring method and device based on a high-resolution satellite image shadow. The method comprises the following steps: extracting a fan shadow by using a shadow extraction model based on a high-resolution satellite image containing a fan; the method comprises the following steps of: segmenting and removing a shadow mask of a fan blade by combining solar azimuth information with a straight line-by-line scanning method, and retaining a shadow mask of a fan tower; and calculating the shadow length of the fan tower according to the shadow mask of the fan tower, and calculating the fan height according to the shadow length of the fan tower. According to the invention, the fan height information can be rapidly obtained in a large range, the fan height can be accurately measured and calculated, and manpower and cost can be saved.

Description

Technical Field

[0001] The present invention relates to the field of remote sensing technology, and in particular to a method and device for measuring the height of a wind turbine based on shadows of high-resolution satellite images. Background Art

[0002] Wind energy is an environmentally friendly, sustainable, and unlimited renewable energy source. Wind power generation has become a crucial component of the global energy transition and is crucial for sustainable development. Measuring wind turbine height is crucial for the safety, efficiency, and maintenance of wind power generation. Accurate height measurement optimizes wind turbine layout, improves wind energy utilization, and enhances power generation efficiency. Furthermore, accurate data helps assess wind turbine structural stability and prevent safety hazards. During operation and maintenance, knowing wind turbine height can aid in developing appropriate maintenance plans, improving operational safety and efficiency. Furthermore, measuring wind turbine height before and after disasters allows for rapid monitoring of turbine status. Therefore, wind turbine height measurement is crucial for ensuring the long-term stable operation of wind farms. Measuring wind turbine height is crucial for optimizing wind energy development, improving wind farm design, and assessing the economic benefits of wind power projects.

[0003] Currently, the main methods for measuring wind turbine height include laser ranging, drone measurement, total station measurement, and LiDAR scanning. However, because wind turbines are often located in areas with complex terrain and variable weather conditions, these field measurement methods are often limited by ground conditions, weather conditions, measurement costs, and acquisition difficulties. These factors can affect the accuracy and feasibility of measurements in different application scenarios. The development of remote sensing technology has provided new means for wind turbine height measurement, especially the use of high-resolution satellite imagery, which can efficiently obtain wind turbine height information over a large area. Summary of the Invention

[0004] In order to solve the above technical problems, the present invention provides a method, device and equipment for extracting building information based on monocular side-view satellite images.

[0005] In a first aspect, a method for measuring wind turbine height based on shadows of high-resolution satellite images is provided, the method comprising: acquiring a high-resolution satellite image; wherein the high-resolution satellite image includes a wind turbine image; extracting the wind turbine shadow from the wind turbine image using a shadow extraction model to obtain a wind turbine shadow mask; wherein the wind turbine shadow mask is composed of pixel points; scanning the wind turbine shadow mask line by line using a scanning line generated by solar azimuth information, segmenting and removing the wind turbine blade shadow mask by determining the number of pixel points at the intersection of the scanning line and the wind turbine shadow mask, thereby retaining the shadow mask of the wind turbine tower; calculating the shadow length of the wind turbine tower based on the shadow mask of the wind turbine tower, and calculating the wind turbine height based on the wind turbine tower shadow length; wherein the wind turbine height is determined by the height of the wind turbine tower.

[0006] In some embodiments, a wind turbine identification model is used to filter out images containing wind turbines from the high-resolution satellite images.

[0007] In some embodiments, the wind turbine recognition model introduces the CBAM attention mechanism based on the YOLOv5 network to enhance the effective feature information of the wind turbine target.

[0008] In some embodiments, the shadow extraction model adopts a U-Net shadow extraction model, which adopts a symmetrical encoder-decoder structure. By introducing multiple jump connections, the feature maps of each stage of the encoder are spliced ​​with the feature maps of the corresponding stage of the decoder to enhance the ability to restore the details of the segmentation boundary.

[0009] In some embodiments, a method for segmenting and removing the shadow mask of a wind turbine blade, thereby retaining the shadow mask of a wind turbine tower, specifically comprises: generating a scanning line along the direction of the solar azimuth angle with a length greater than the wind turbine shadow mask, the scanning line scanning from the left side of the tower to the right side along a direction perpendicular to the wind turbine shadow mask; recording the number of pixels at which the wind turbine blade shadow first overlaps with the scanning line, and continuing to scan to the right; when the number of pixels at which the scanning line overlaps with the wind turbine blade shadow is greater than or equal to a first preset threshold, using the scanning line at that position as a segmentation line to segment and remove the wind turbine blade shadow on the left side of the scanning line; wherein the first preset threshold is determined based on a fixed width ratio between the end and the root of the wind turbine blade; the scanning line is generated along a direction perpendicular to the wind turbine shadow mask The wind turbine shadow mask is scanned from the right side to the left side of the tower in the membrane direction, and the wind turbine blade shadow on the right side of the scanning line is segmented and removed in the manner described in the previous step; a scanning line with a length greater than the wind turbine shadow mask is generated in the direction perpendicular to the solar azimuth, and the scanning line is scanned from the top to the bottom of the tower along the direction parallel to the wind turbine mask; the number of pixel points at which the wind turbine blade shadow coincides with the scanning line for the first time is recorded, and the scanning is continued toward the bottom of the tower. When the number of pixel points at which the scanning line coincides with the wind turbine cabin shadow is greater than or equal to a second preset threshold, the scanning line at that position is used as a dividing line to segment and remove the wind turbine blade shadow at the top of the tower; wherein the second preset threshold is determined based on the fixed width ratio between the end of the wind turbine blade and the wind turbine cabin.

[0010] In some embodiments, a method for calculating the shadow length of a wind tower based on the shadow mask of the wind tower specifically includes: obtaining the shadow length values ​​of multiple wind towers by calculating the number of pixels along the solar azimuth angle within the shadow range of the wind tower; using the Laida criterion to eliminate the error values ​​of the shadow length of the wind tower, and recalculating the average shadow length of the wind tower.

[0011] In some embodiments, the method for calculating the height of a wind turbine based on the shadow length of the wind turbine tower specifically includes:

[0012] When the azimuth angle between the sun and the satellite is greater than 180°, the height of the wind turbine is calculated using the following formula:

[0013] H=S×tanα

[0014] Where H is the height of the wind turbine, S is the length of the wind turbine shadow observed on the remote sensing image, and α is the solar altitude angle;

[0015] When the sun and satellite are in the same orientation, the height of the wind turbine is calculated using the following formula:

[0016]

[0017] Where H is the height of the wind turbine, S is the length of the wind turbine shadow observed on the remote sensing image, α is the solar altitude angle, and β is the satellite altitude angle;

[0018] When the azimuth angle difference between the sun and the satellite is between 0° and 180°, the height of the wind turbine is calculated using the following formula:

[0019]

[0020] Where H is the height of the wind turbine, S is the length of the wind turbine shadow observed on the remote sensing image, γ is the solar azimuth, δ is the satellite azimuth, and ε is the angle between the wind turbine direction and its shadow projection in the clockwise direction.

[0021] In a second aspect, a device for measuring wind turbine height based on shadows of high-resolution satellite images is provided according to the method of the first aspect, the device comprising:

[0022] A wind turbine identification module acquires high-resolution satellite images; wherein the high-resolution satellite images include wind turbine images;

[0023] a fan shadow extraction module, configured to extract the fan shadow from the fan image using a shadow extraction model to obtain a fan shadow mask; wherein the fan shadow mask is composed of pixel points;

[0024] a wind turbine blade shadow removal module, configured to scan the wind turbine shadow mask line by line using a scanning line generated by the solar azimuth information, and segment and remove the wind turbine blade shadow mask by determining the number of pixels at the intersection of the scanning line and the wind turbine shadow mask, thereby retaining the wind turbine tower shadow mask;

[0025] The wind turbine height calculation module is used to calculate the shadow length of the wind turbine tower according to the shadow mask of the wind turbine tower, and calculate the wind turbine height according to the shadow length of the wind turbine tower; wherein the wind turbine height is determined by the height of the wind turbine tower.

[0026] Traditional wind turbine height measurement methods are affected by factors such as terrain and weather, resulting in high costs and low efficiency. Furthermore, field measurements make it difficult to cover large wind farms. The present invention, based on high-resolution satellite imagery containing wind turbine images, uses a shadow extraction model to extract wind turbine shadows, and utilizes solar azimuth information combined with a linear, line-by-line scanning method to eliminate blade shadow interference and improve measurement accuracy. Combining solar altitude and azimuth parameters, a wind turbine height inversion model is constructed to achieve automated, high-precision measurement. This method can quickly acquire wind turbine height information over a large area, accurately calculate wind turbine height, and save manpower and costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0028] Figure 1 A schematic flow chart of a method for measuring wind turbine height based on shadows of high-resolution satellite images provided in an embodiment of this specification is shown;

[0029] Figure 2 A schematic diagram of a wind turbine identification model architecture provided in an embodiment of this specification is shown;

[0030] Figure 3 A schematic diagram of a U-net shadow extraction model architecture provided in an embodiment of this specification is shown;

[0031] Figure 4 A schematic flow chart of a method for segmenting and removing a shadow mask of a wind turbine blade provided in an embodiment of this specification is shown;

[0032] Figure 5 A schematic diagram of wind turbine blade shadow mask segmentation and removal provided by an embodiment of this specification is shown;

[0033] Figure 6 A schematic structural diagram of a wind turbine height measurement device based on high-resolution satellite image shadows provided in an embodiment of this specification is shown. DETAILED DESCRIPTION

[0034] The solution provided in this specification is described below in conjunction with the accompanying drawings.

[0035] 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 described below with reference to the accompanying drawings.

[0036] In the description of the embodiments of the present application, words such as "exemplary," "for example," or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary," "for example," or "for example" in the embodiments of the present application should not be construed as being preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary," "for example," or "for example" is intended to present the relevant concepts in a concrete manner.

[0037] In the description of the embodiments of this application, the term "and / or" is merely a description of an association relationship between associated objects, indicating that three relationships may exist. For example, A and / or B can represent the following three situations: A exists alone, B exists alone, and A and B exist at the same time. In addition, unless otherwise specified, the term "plurality" means two or more.

[0038] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly identifying the technical features being referred to. Thus, features specified as "first" or "second" may explicitly or implicitly include one or more of such features. The terms "include," "comprising," "having," and their variations all mean "including but not limited to," unless otherwise specifically emphasized.

[0039] Figure 1 The following is a flow chart of a method for measuring wind turbine height based on high-resolution satellite image shadows provided in an embodiment of this specification. Figure 1 As shown, the method includes the following steps:

[0040] Step S101: Acquire high-resolution satellite images, wherein the high-resolution satellite images include wind turbine images.

[0041] Specifically, high-resolution satellite images can be obtained from official channels or authorized platforms. Preferably, Gaofen-2 satellite images are selected. High-resolution satellite images include wind turbine images.

[0042] In some embodiments, a wind turbine recognition model can be used to filter out images containing wind turbines from high-resolution satellite images. The wind turbine recognition model can be selected from CNNs, R-CNN, YOLO, Mask R-CNN, etc.

[0043] In some specific embodiments, the wind turbine identification model can use the YOLOv5-CBAM model, and its specific architecture is as follows: Figure 2 The YOLOv5-CBAM model introduces the CBAM attention mechanism based on the YOLOv5 network to enhance the effective feature information of the wind turbine target.

[0044] The YOLOv5 algorithm is a real-time, end-to-end object detection method. It considers the importance of wind turbine target pixels in different channels and at different spatial locations in the same channel, and introduces the CBAM attention mechanism based on the existing YOLOv5 network.

[0045] Preferably, three CBAM modules are provided, of which the first CBAM module can be placed in the backbone network, more specifically, before the first upsampling module. The other two CBAM modules can be placed in the feature pyramid of the network's Neck portion, specifically, before the last two feature concat modules. These modules can be used to enhance the effective feature information of the wind turbine target and weaken the invalid feature information, thereby improving the network model extraction capability.

[0046] After the image passes through the backbone network for feature extraction, the different channels of the feature map contain multi-dimensional feature information of the wind turbine target, enhancing the wind turbine target features in important channels, thereby improving the discrimination performance of the wind turbine target. In the spatial dimension, CBAM can generate a spatial attention feature map based on spatial relationships. By considering the spatial relationships between pixels, it enhances the key features of the wind turbine target in the spatial dimension and improves the model's ability to mine the semantic information of the wind turbine target.

[0047] In some embodiments, before screening images containing wind turbines, the high-resolution satellite images can be preprocessed. Specifically, radiometric and geometric corrections are performed on the panchromatic image within the high-resolution satellite image, while radiometric, atmospheric, and geometric corrections are performed on the multispectral image. The preprocessed panchromatic and multispectral images are then fused to obtain a preprocessed, usable high-resolution satellite image.

[0048] In step S102 , a shadow extraction model may be used to extract the wind turbine shadow from the wind turbine image to obtain a wind turbine shadow mask.

[0049] You can collect images containing wind turbine shadows, and then select a suitable shadow extraction model, such as DeepLab, SegNet, PSPNet, U-Net and other models. Finally, use the selected shadow extraction model to extract the wind turbine shadow from the image containing the wind turbine shadow, thereby obtaining the wind turbine shadow mask.

[0050] The wind turbine shadow mask refers to a digitized layer that identifies the wind turbine shadow area. In other words, the wind turbine shadow mask is composed of pixels.

[0051] In some embodiments, the shadow extraction model can adopt a U-Net shadow extraction model. Figure 3 As shown in , its U-Net shadow extraction model can adopt a symmetrical encoder-decoder structure.

[0052] Specifically, refer to Figure 3, multiple skip connections (referred to as Copy and Crop operations in the figure) can be introduced to directly concatenate the feature maps of each encoder stage with the feature maps of the corresponding decoder stage, thereby enhancing the ability to recover detailed information at segmentation boundaries. The encoding path primarily consists of continuous convolution operations and max pooling layers, gradually compressing spatial information and extracting features; the decoding path restores the spatial scale through deconvolution (or upsampling) and convolution operations, and integrates local information from the encoder.

[0053] Through the above structure, U-Net can take into account both local details and global context features with limited labeled data and achieve accurate pixel-level segmentation.

[0054] In step S103, the wind turbine shadow mask is scanned line by line using a scanning line generated by the solar azimuth information. The wind turbine blade shadow mask is segmented and removed by determining the number of pixels at the intersection of the scanning line and the wind turbine shadow mask, thereby retaining the wind turbine tower shadow mask.

[0055] Specifically, wind turbine height refers to the height of the wind turbine tower. For height inversion, only the tower's shadow length is required. When automatically calculating the wind turbine tower's shadow length, the blades' shadows can affect the calculated length, causing errors in the result. To address this issue, we use line-by-line scanning combined with solar azimuth angle information to remove blade shadows from the wind turbine, ensuring accurate tower shadow length calculations.

[0056] In some embodiments, Figure 4 A method for segmenting and removing the shadow mask of the wind turbine blades, thereby retaining the shadow mask of the wind turbine tower, is shown. Figure 4 As shown, the method may include:

[0057] Step S1031 : generating a scanning line with a length greater than the wind turbine shadow mask along the direction of the solar azimuth angle, and scanning the scanning line from the left side to the right side of the tower along a direction perpendicular to the wind turbine shadow mask.

[0058] It's easy to understand that the solar azimuth is parallel to the axis of the wind turbine shadow mask, so the scanning line generated along the solar azimuth is parallel to the axis of the wind turbine blade shadow mask. This scanning line can be used to scan from the left side of the tower to the right side, perpendicular to the wind turbine shadow mask.

[0059] Step S1032 records the number of pixels where the fan blade shadow first overlaps with the scanning line, and continues scanning to the right. When the number of pixels where the scanning line overlaps with the fan blade shadow is greater than or equal to a first preset threshold, the scanning line at that position is used as a dividing line to segment and remove the fan blade shadow to the left of the scanning line.

[0060] Specifically, different solar azimuths or altitudes will change the overall shape of the wind turbine shadow, but will not change the width ratio between the tip and root of the wind turbine blade. The first preset threshold can be determined based on the fixed width ratio between the tip and root of the wind turbine blade.

[0061] For example, the width ratio of the end and root of a fan blade is fixed at 1:7. When the number of pixels at the end of the fan blade shadow is 2, the first preset threshold can be set to 14. That is, when the number of pixels where the scanning line overlaps with the fan blade shadow is greater than or equal to the first preset threshold 14, it can be determined that the position is the root of the fan blade. Then the scanning line at this position can be used as a dividing line, and then the fan blade shadow on the left side of the scanning line can be segmented and removed. The fan shadow mask before the fan blade shadow segmentation and removal can be referred to. Figure 5 (a) The effect of fan blade shadow segmentation and removal on the left side of the scanning line can be referred to Figure 5 (b).

[0062] Step S1033, scan the fan shadow mask from the right side of the tower to the left side along the scanning line perpendicular to the fan shadow mask direction, and segment and remove the fan blade shadow on the right side of the scanning line in the same manner as described in the previous step. The difference between this step and the previous step is the direction of the linear scanning. The implementation methods involved in the other steps are similar and will not be repeated here. The effect of segmenting and removing the fan blade shadow on the right side of the scanning line can be referred to Figure 5 (c).

[0063] Step S1034: Generate a scanning line with a length greater than the wind turbine shadow mask in a direction perpendicular to the solar azimuth angle. The scanning line scans from the top of the tower to the bottom in a direction parallel to the wind turbine mask.

[0064] It's easy to understand that the solar azimuth is parallel to the axis of the wind turbine's shadow mask. Based on geometric principles, a line drawn perpendicular to the solar azimuth is also perpendicular to the axis of the wind turbine's shadow mask. This scanning line can be used to scan from the top of the tower to the bottom, parallel to the wind turbine's shadow mask.

[0065] Step S1035, record the number of pixel points where the wind blade shadow coincides with the straight line for the first time, continue scanning toward the bottom of the tower, and when the number of pixel points where the scanning straight line coincides with the wind turbine cabin shadow is greater than or equal to the second preset threshold, use the scanning straight line at that position as a dividing line to divide and remove the wind blade shadow at the top of the tower.

[0066] Specifically, the wind turbine nacelle is part of the tower and is located at the top of the tower. The shadow where the three wind turbine blades meet the nacelle is wider than the shadow where the blade roots meet the tower. Because the nacelle is located at the top of the tower, the top of the tower's shadow should align with the top of the nacelle's shadow. In other words, the top of the nacelle's shadow is also the top of the tower's shadow.

[0067] Different solar azimuths or altitudes will change the overall shape of the wind turbine shadow, but will not change the width ratio between the wind turbine blade ends and the wind turbine nacelle. The second preset threshold can be determined based on the fixed width ratio between the wind turbine blade ends and the wind turbine nacelle.

[0068] For example, the width ratio of the end of the wind turbine blade and the wind turbine cabin is 1:10. When the pixel point of the end of the wind turbine blade shadow is 1, the second preset threshold can be set to 10. That is to say, when the number of pixels where the scanning line coincides with the wind turbine cabin shadow is greater than or equal to the second preset threshold 10, it can be determined that the position is the top of the tower, and then the scanning line at this position can be used as the dividing line (such as Figure 5 (d)), and then finally segment and remove the shadow of the wind turbine blades at the top of the tower. The effect of segmenting and removing the shadow of the wind turbine blades at the top of the tower can be referred to Figure 5 (e).

[0069] It can be seen from the above embodiments that the method provided in this example is used to segment and remove the shadow of the wind turbine blades, which can solve the problem that the shadow of the wind turbine blades will affect the length calculation of the tower shadow and cause errors in the result when the shadow length is automatically calculated, and can improve the calculation accuracy of the wind turbine height.

[0070] Step S104: Calculate the shadow length of the wind turbine tower according to the shadow mask of the wind turbine tower, and calculate the wind turbine height according to the shadow length of the wind turbine tower, wherein the wind turbine height is determined by the height of the wind turbine tower.

[0071] Specifically, the minimum bounding rectangle method, solar geometry correction method, anti-interference optimization method, skeletonization method, pixel counting method, etc. can be used to calculate the shadow length of the wind turbine tower based on the shadow mask of the wind turbine tower. The wind turbine height can be calculated based on the azimuth and altitude angles of the sun and satellite and the location of the wind turbine.

[0072] In one embodiment, the shadow length of the wind turbine tower may be calculated using the pixel counting method and the Laida criterion.

[0073] Specifically, by calculating the number of pixels along the solar azimuth within the shadow range of the wind tower, the shadow length values ​​of multiple wind towers are obtained. The Laida criterion is then used to eliminate the errors in the shadow length values ​​of the wind towers and recalculate the average wind tower shadow length.

[0074] For example, the process of eliminating the error value of wind turbine tower shadow length using the Laida criterion is as follows:

[0075] (1) Calculate the standard deviation σ and arithmetic mean μ of the tower shadow length obtained.

[0076] (2) Calculate the deviation of the length of each overlapping line segment relative to the mean μi as follows:

[0077] x i =|μ i -μ|

[0078] (3) Taking 3σ as the normal value range, the corresponding confidence level is 99.73%.

[0079] (4) If x i ≤3σ, the length of the overlapping line segment is considered normal and retained; if x i If the value is >3σ, it is considered as an outlier and eliminated.

[0080] (5) Since removing outliers will change the sample mean and standard deviation, this process needs to be repeated until no more data points are excluded.

[0081] In some embodiments, the wind turbine height can be inverted using the wind turbine tower shadow based on the azimuth relationship between the sun and the satellite. This includes situations where the azimuth difference between the sun and the satellite is greater than 180°, when the azimuths of the sun and the satellite are the same, and when the azimuth difference between the sun and the satellite is between 0° and 180°.

[0082] When the azimuth angle between the sun and the satellite is greater than 180°, the satellite is located on the opposite side of the wind turbine from the sun when acquiring surface information. The satellite can observe all shadow areas of the wind turbine. The formula for calculating the height of the wind turbine is as follows:

[0083] H=S×tanα

[0084] Where H is the height of the wind turbine, S is the length of the wind turbine shadow observed on the remote sensing image, and α is the solar altitude angle.

[0085] When the sun and satellite are in the same orientation, the satellite and the sun are on the same side of the wind turbine. In this case, the influence of the azimuth angle does not need to be considered, and part of the shadow is blocked by the wind turbine itself. In this case, the formula for calculating the wind turbine height is as follows:

[0086]

[0087] Where H is the height of the wind turbine, S is the length of the wind turbine shadow observed on the remote sensing image, α is the solar altitude angle, and β is the satellite altitude angle.

[0088] When the azimuth angle difference between the sun and the satellite is between 0° and 180°, the wind turbine height calculation formula is as follows:

[0089]

[0090] Where H is the height of the wind turbine, S is the length of the wind turbine shadow observed on the remote sensing image, γ is the solar azimuth, δ is the satellite azimuth, and ε is the angle between the wind turbine direction and its shadow projection in the clockwise direction.

[0091] Corresponding to the above method provided by the present invention, the present invention also provides a device. Figure 6 FIG. 1 shows a schematic diagram of a wind turbine height measurement device based on high-resolution satellite image shadows provided by an embodiment of this specification. Figure 6 As shown, the device includes:

[0092] The wind turbine identification module 201 is used to obtain high-resolution satellite images, wherein the high-resolution satellite images include wind turbine images.

[0093] The wind turbine shadow extraction module 202 is configured to extract the wind turbine shadow from the wind turbine image using a shadow extraction model to obtain a wind turbine shadow mask, wherein the wind turbine shadow mask is composed of pixels.

[0094] The wind turbine blade shadow removal module 203 is used to scan the wind turbine shadow mask line by line using a scanning line generated by the solar azimuth information, and to segment and remove the wind turbine blade shadow mask by determining the number of pixels at the intersection of the scanning line and the wind turbine shadow mask, thereby retaining the wind turbine tower shadow mask.

[0095] The wind turbine height calculation module 204 is configured to calculate the length of the wind turbine tower's shadow based on the wind turbine tower's shadow mask, and calculate the wind turbine height based on the wind turbine tower's shadow length; wherein the wind turbine height is determined by the height of the wind turbine tower.

[0096] It is worth mentioning that Figure 6 The device shown is Figure 1 The method embodiments shown correspond to and can be applied to Figure 1The high-resolution satellite image acquisition, wind turbine shadow extraction, wind turbine blade shadow removal, and wind turbine height calculation in the illustrated method embodiment will not be described in detail here. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of each example described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in hardware or in a computer software-driven hardware manner depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0097] According to another embodiment, a computing device is provided, including a memory and a processor, wherein the memory stores executable code, and when the processor executes the executable code, the system realizes the combination of Figure 1 The method described.

[0098] According to another embodiment, there is also provided a computer readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to execute a combination of Figure 1 The method described.

[0099] Those skilled in the art will appreciate that, in one or more of the above examples, the functions described herein may be implemented using hardware, software, firmware, or any combination thereof. When implemented using software, these functions may be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium.

[0100] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent replacements, improvements, etc. made on the basis of the technical solution of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for measuring wind turbine height based on high-resolution satellite image shadows, characterized in that: The method comprises: Acquire a high-resolution satellite image; wherein the high-resolution satellite image includes a wind turbine image; Extracting the wind turbine shadow from the wind turbine image using a shadow extraction model to obtain a wind turbine shadow mask; wherein the wind turbine shadow mask is composed of pixel points; Scanning the wind turbine shadow mask line by line using a scanning line generated by solar azimuth information, and segmenting and removing the wind turbine blade shadow mask by determining the number of pixels at intersections between the scanning line and the wind turbine shadow mask, thereby retaining the wind turbine tower shadow mask; The shadow length of the wind turbine tower is calculated according to the shadow mask of the wind turbine tower, and the wind turbine height is calculated according to the shadow length of the wind turbine tower; wherein the wind turbine height is determined by the height of the wind turbine tower.

2. The method according to claim 1, characterized in that The wind turbine identification model is used to filter out images containing wind turbines from the high-resolution satellite images.

3. The method according to claim 2, characterized in that The wind turbine recognition model introduces the CBAM attention mechanism based on the YOLOv5 network to enhance the effective feature information of the wind turbine target.

4. The method according to claim 1, wherein The shadow extraction model adopts the U-Net shadow extraction model, which adopts a symmetrical encoder-decoder structure. By introducing multiple jump connections, the feature maps of each stage of the encoder are spliced ​​with the feature maps of the corresponding stage of the decoder to enhance the ability to restore details of the segmentation boundary.

5. The method according to claim 1, characterized in that The method for segmenting and removing the shadow mask of the wind turbine blades so as to retain the shadow mask of the wind turbine tower specifically includes: A scanning line with a length greater than the wind turbine shadow mask is generated along the direction of the solar azimuth angle. The scanning line is scanned from the left side to the right side of the tower along the direction perpendicular to the wind turbine shadow mask. Record the number of pixels where the fan blade shadow first overlaps with the scanning line, and continue scanning to the right. When the number of pixels where the scanning line overlaps with the fan blade shadow is greater than or equal to a first preset threshold, use the scanning line at that position as a dividing line to divide and remove the fan blade shadow to the left of the scanning line. The first preset threshold is determined based on a fixed width ratio between the end and root of the fan blade. Scan the fan shadow mask from the right side of the tower to the left side along a scanning line perpendicular to the fan shadow mask. Segment and remove the fan blade shadows on the right side of the scanning line in the same manner as described in the previous step. Generate a scanning line with a length greater than the wind turbine shadow mask in a direction perpendicular to the solar azimuth angle. The scanning line scans from the top of the tower to the bottom in a direction parallel to the wind turbine mask. Record the number of pixels where the wind blade shadow first overlaps with the scanning line, and continue scanning toward the bottom of the tower. When the number of pixels where the scanning line overlaps with the wind turbine nacelle shadow is greater than or equal to a second preset threshold, use the scanning line at that position as a dividing line to divide and remove the wind blade shadow at the top of the tower. The second preset threshold is determined based on a fixed width ratio between the end of the wind blade and the wind turbine nacelle.

6. The method according to claim 1, characterized in that The method for calculating the shadow length of a wind turbine tower according to the shadow mask of the wind turbine tower specifically includes: By calculating the number of pixels along the solar azimuth angle within the shadow range of the wind turbine tower, the shadow length values ​​of multiple wind turbine towers are obtained; The Laida criterion is used to eliminate the error value of the wind turbine tower shadow length and recalculate the average wind turbine tower shadow length.

7. The method according to claim 1, characterized in that The method for calculating the height of the wind turbine according to the shadow length of the wind turbine tower specifically includes: When the azimuth angle between the sun and the satellite is greater than 180°, the height of the wind turbine is calculated using the following formula: H=S×tanα Where H is the height of the wind turbine, S is the length of the wind turbine shadow observed on the remote sensing image, and α is the solar altitude angle; When the sun and satellite are in the same orientation, the height of the wind turbine is calculated using the following formula: Where H is the height of the wind turbine, S is the length of the wind turbine shadow observed on the remote sensing image, α is the solar altitude angle, and β is the satellite altitude angle; When the azimuth angle difference between the sun and the satellite is between 0° and 180°, the height of the wind turbine is calculated using the following formula: Where H is the height of the wind turbine, S is the length of the wind turbine shadow observed on the remote sensing image, γ is the solar azimuth, δ is the satellite azimuth, and ε is the angle between the wind turbine direction and its shadow projection in the clockwise direction.

8. A wind turbine height measurement device based on high-resolution satellite image shadows, characterized in that: The wind turbine height is measured by the method according to any one of claims 1 to 7, wherein the device comprises: A wind turbine identification module acquires high-resolution satellite images; wherein the high-resolution satellite images include wind turbine images; a fan shadow extraction module, configured to extract the fan shadow from the fan image using a shadow extraction model to obtain a fan shadow mask; wherein the fan shadow mask is composed of pixel points; a wind turbine blade shadow removal module, configured to scan the wind turbine shadow mask line by line using a scanning line generated by the solar azimuth information, and segment and remove the wind turbine blade shadow mask by determining the number of pixels at the intersection of the scanning line and the wind turbine shadow mask, thereby retaining the wind turbine tower shadow mask; The wind turbine height calculation module is used to calculate the shadow length of the wind turbine tower according to the shadow mask of the wind turbine tower, and calculate the wind turbine height according to the shadow length of the wind turbine tower; wherein the wind turbine height is determined by the height of the wind turbine tower.

Images