A method and system for measuring wind field based on sky polarization background movement

Abstract

This invention discloses a wind field measurement method and system based on the movement of the sky polarization background. The ingenious approach involves capturing images of a preset area of ​​the sky from multiple angles. Then, the target object in the sky image is located, selected, and its position (within a preset reference or virtual coordinate system) is reconstructed by combining the position information of the image acquisition device. Furthermore, the trajectory is determined by analyzing the positional changes of the target object in different groups of sky images divided according to time points, thereby retrieving the wind field and obtaining further wind field information. This solution facilitates wind field information estimation, reduces equipment requirements, and leverages the computing power of a server to post-process the sky images obtained by the image acquisition device, achieving a high-speed response feedback effect. This provides a new approach and direction for wind field measurement.

Description

Technical Field

[0001] This invention relates to the field of wind field measurement technology, and in particular to a wind field measurement method and system based on the movement of the sky polarization background. Background Technology

[0002] Atmospheric wind field measurement plays an important role in aerospace, pollutant dispersion and other assessments. At present, the main methods of atmospheric wind field measurement are through sonar, microwave radar and lidar. However, these measurement devices often require a large investment of funds and materials. Therefore, although they are relatively accurate, they still have significant limitations in terms of flexibility, measurement area switching and low cost.

[0003] With the increasing maturity of image acquisition and photo fitting technologies, numerous researchers have published studies on locating the 3D coordinates of identical elements within multiple images. Handheld devices equipped with high-definition image acquisition modules, such as mobile phones and tablets, are now widely available. Furthermore, existing handheld terminals are equipped with numerous sensors to allow users to easily check their current location and handheld device orientation. If image acquisition devices could be used to obtain the polarized background of the sky, and the atmospheric wind field could be estimated based on the time difference obtained from combining multiple images, this would greatly facilitate atmospheric wind field measurement, improve the operational flexibility of testing personnel, increase data acquisition efficiency, and reduce testing costs. Summary of the Invention

[0004] In view of this, the purpose of this invention is to provide a reliable, convenient and flexible wind field measurement method and system based on the movement of the sky polarization background.

[0005] To achieve the above-mentioned technical objectives, the technical solution adopted by this invention is as follows:

[0006] A wind field measurement method based on sky polarization background movement, comprising:

[0007] S01. Multiple sky images are obtained by capturing images of a preset area of ​​the sky from different shooting angles using multiple image capturing devices equipped with polarizing filters.

[0008] S02. The obtained sky images are labeled with information and then stored to generate a sky image set; wherein, the information labeling content includes at least the time point when the sky image was captured and the location information of the image capturing device that captured the sky image;

[0009] S03. Determine the target object in the preset area of ​​the sky according to the preset conditions;

[0010] S04. Match the sky images in the sky image set according to the characteristics of the target object to obtain multiple sky images with the target object. Then, group the sky images with the same shooting time into a group to obtain at least two groups of sky images. Each group of sky images contains at least three sky images, and the corresponding image capturing devices are different.

[0011] S05. Determine the location information of the target object in the same set of sky images based on at least three sky images in the same set and the location information of the image capturing device associated with them;

[0012] S06. Based on the shooting time points and the location information of the target object corresponding to at least two sets of sky images, the wind field in the area where the target object is located is inverted to obtain wind field information.

[0013] As one possible implementation, further, in this solution S01, each image capturing device has two or more cameras, and the two or more cameras take pictures simultaneously with not exactly the same focusing parameters. The captured photos are processed according to preset conditions to generate a sky image.

[0014] Among them, the image capture range of multiple image capturing devices completely covers the preset area of ​​the sky;

[0015] In addition, among the multiple image capturing devices, there is at least one other image capturing device whose image capturing range overlaps with that of one of the image capturing devices by 30% to 70%.

[0016] As a preferred implementation method, in S01 of this solution, two or more cameras of the image capturing device automatically focus on a preset area of ​​the sky using a multi-point focusing method to obtain focus parameters and take pictures. When the focus parameters of two or more cameras are the same, the focusing is re-executed so that the focus parameters of different cameras are not completely the same, and the two or more cameras take pictures synchronously with not completely the same focus parameters to obtain multiple sky images that meet the preset requirements and upload them to the server.

[0017] As a preferred implementation method, in S02 of this solution, before annotating the obtained sky image, the elements with non-sky solid color backgrounds in the images captured by two or more cameras of the image capturing device are located and selected. Then, sky images with only sky solid color backgrounds and interfering elements are removed. The remaining sky images are then judged for clarity. Finally, the image with the highest element clarity is retained, so that each element with a non-sky solid color background in the sky image of the same sky area corresponds to a sky image, generating at least one sky image.

[0018] As a preferred implementation method, in S02 of this solution, when annotating the obtained sky image, the elements in the sky image that are not in a solid color background are also annotated with outlines, and reference points are further marked on the elements according to the outline annotations. The number of reference points is at least one. When there is one reference point, it is the geometric center of the outline annotation. When there is more than one reference point, at least one reference point is located at the geometric center of the outline annotation.

[0019] As a preferred implementation method, preferably, in this solution S03, the method of determining the target object in the preset area of ​​the sky according to preset conditions includes: specifying the target object in the sky image that meets the preset conditions; when the specified target object has a contour annotation in the information annotation of the sky image, it is determined that the target object meets the requirements, and then its features are extracted.

[0020] As a preferred implementation method, preferably, in this solution S03, the edge of the target object is completely located in the sky image, and its occupancy ratio in the sky image is greater than 5% and less than 40%.

[0021] As a preferred implementation method, the target object described in this solution is something that floats in the sky solely due to its own gravity and the force of the wind, including clouds, kites without strings, balloons without strings, and leaves.

[0022] As a preferred implementation method, preferably, in this solution S01, the polarizing mirror mounted on the image capturing device is used to filter stray light in the shooting environment.

[0023] Based on the above, the present invention also provides the application of a handheld terminal as an image acquisition device in atmospheric wind field measurement, which includes the wind field measurement method based on the movement of the sky polarization background described above.

[0024] The handheld terminal is a mobile phone, handheld tablet computer, or camera equipped with a communication module, an image capture module, a pose sensor, and a positioning module.

[0025] In addition, a polarizing filter is provided on the lens of the image acquisition module.

[0026] Based on the above, the present invention also provides a wind field measurement system based on the movement of the sky polarization background, which includes:

[0027] The image acquisition devices are multiple and each is equipped with a polarizing filter. The multiple image acquisition devices capture images of a preset area of ​​the sky from different shooting angles to obtain multiple sky images.

[0028] Storage unit, used to store sky images captured by the image capturing device;

[0029] An information annotation unit is used to annotate the acquired sky images and then store them to generate a sky image set; wherein, the content of the information annotation includes at least the time point when the sky image was captured and the location information of the image capturing device that captured the sky image;

[0030] The target determination unit is used to determine the target object in a preset area of ​​the sky according to preset conditions.

[0031] The data matching unit is used to match the sky image set according to the characteristics of the target object to obtain multiple sky images with the target object. Then, the sky images with the same shooting time are grouped into a group to obtain at least two groups of sky images. Each group of sky images contains at least three sky images, and the corresponding image capturing devices are different.

[0032] The target positioning unit is used to determine the location information of a target object in the same set of sky images based on at least three sky images in the same set and the location information of the image acquisition device associated with them.

[0033] The data processing unit is used to invert the wind field in the area where the target object is located based on the shooting time point corresponding to at least two sets of sky images and the location information of the target object, thereby obtaining wind field information.

[0034] Based on the above, the present invention also provides a computer-readable storage medium, characterized in that: the storage medium stores at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, at least one program, code set, or instruction set is loaded and executed by a processor to implement the wind field measurement method based on the movement of the sky polarization background described above.

[0035] Compared with the prior art, the present invention has the following advantages by adopting the above technical solution: The present invention ingeniously captures images of a preset area of ​​the sky through multi-angle image acquisition, then locates and selects the target object in the sky image, and combines the position information of the image acquisition device to restore the position of the target object (position under a preset reference or virtual coordinate system). Then, by judging the position change of the target object in different groups of sky images divided according to time points, its trajectory is determined, thereby reversing the wind field and further obtaining wind field information. The present invention reduces the equipment requirements while facilitating wind field information estimation. At the same time, by using the computing power of the server to post-process the sky image obtained by the image acquisition device, a high-speed response feedback effect can be achieved, providing a new idea and direction for wind field measurement. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a simplified implementation flowchart of the wind field measurement method of the present invention;

[0038] Figure 2 This is a simplified schematic diagram of the present invention, which uses a handheld terminal as an image capturing device to acquire sky images.

[0039] Figure 3 This is a simplified schematic diagram of the modular unit connections of the system according to the present invention. Detailed Implementation

[0040] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be particularly noted that the following embodiments are for illustrative purposes only and do not limit the scope of the invention. Similarly, the following embodiments are only some, not all, embodiments of the present invention, and all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0041] Combination Figure 1 As shown, a wind field measurement method based on sky polarization background movement includes:

[0042] S01. Multiple sky images are obtained by capturing images of a preset area of ​​the sky from different shooting angles using multiple image capturing devices equipped with polarizing filters.

[0043] S02. The obtained sky images are labeled with information and then stored to generate a sky image set; wherein, the information labeling content includes at least the time point when the sky image was captured and the location information of the image capturing device that captured the sky image;

[0044] S03. Determine the target object in the preset area of ​​the sky according to the preset conditions;

[0045] S04. Match the sky images in the sky image set according to the characteristics of the target object to obtain multiple sky images with the target object. Then, group the sky images with the same shooting time into a group to obtain at least two groups of sky images. Each group of sky images contains at least three sky images, and the corresponding image capturing devices are different.

[0046] S05. Determine the location information of the target object in the same set of sky images based on at least three sky images in the same set and the location information of the image capturing device associated with them;

[0047] S06. Based on the shooting time points and the location information of the target object corresponding to at least two sets of sky images, the wind field in the area where the target object is located is inverted to obtain wind field information.

[0048] In this solution S01, the polarizing mirror mounted on the image capturing device is used to filter stray light in the shooting environment, so as to avoid the sky image captured by the image capturing device being affected by stray light, thereby improving the quality of the captured image.

[0049] Based on this, since there may be more than one reference object (potential target object) in the sky image that can be used as a wind field measurement, and with a single camera, the image acquisition device can only capture one image at the same time point, even with high-speed shooting, there is still a small time difference between the images. In order to obtain the clarity of potential target objects stably, in this solution, each image acquisition device has two or more cameras, and the two or more cameras shoot synchronously with not exactly the same focusing parameters. The shooting frequency is set according to a preset frequency value, such as shooting once, twice, three times, four times, five times per second, etc. The captured photos are processed according to preset conditions to generate sky images.

[0050] Among them, the image capture range of multiple image capturing devices completely covers the preset area of ​​the sky;

[0051] In addition, among the multiple image acquisition devices, there is at least one image acquisition device whose image acquisition range overlaps with that of one of the image acquisition devices by 30% to 70%. By adopting a certain degree of overlap in the image acquisition range, blind spots in the sky area can be avoided. At the same time, when a potential target object moves from the area captured by one or more image acquisition devices to the shooting range of another or more image acquisition devices, its trajectory can be obtained intuitively and conveniently.

[0052] In addition, as a preferred implementation method, in S01 of this solution, two or more cameras of the image capturing device automatically focus on a preset area of ​​the sky using a multi-point focusing method to obtain focus parameters and take pictures. When the focus parameters of two or more cameras are the same, the focusing is re-executed so that the focus parameters of different cameras are not completely the same, and the two or more cameras take pictures synchronously with not completely the same focus parameters to obtain multiple sky images that meet the preset requirements and upload them to the server.

[0053] In this solution, the image acquisition device only participates in the work of capturing sky images and a very small amount of other auxiliary work. This reduces the hardware requirements of the image acquisition device, while the captured sky images can be processed by the back-end server. With the help of the high performance of the cloud or the back-end server, it can provide high processing efficiency and accuracy for subsequent wind field measurements.

[0054] Since the sky background may consist of only a solid color background or no other reference value target objects, in order to save backend storage resources and reduce the workload of subsequent data matching, thereby improving processing efficiency, as a preferred implementation method, in this solution S02, before annotating the information of the obtained sky image, the elements with non-sky solid color backgrounds in the images captured by two or more cameras of the image capturing device are located and selected. Then, sky images with only a solid color sky background and interfering elements (such as airplanes, birds, and other floating objects with a third traction or driving force other than wind field and self-weight) are removed. Then, the clarity of the remaining sky images is judged, and finally, the image with the highest element clarity is retained, so that in the sky images of the same sky area, each element with a non-sky solid color background corresponds to a sky image, generating at least one sky image.

[0055] Among them, the element localization of the sky image, the determination of whether the localized element is a disturbance element, and the determination of whether the sky image is a solid color background can be assisted by a trained neural network. The training method of the neural network has been revealed in existing literature, so it will not be elaborated here.

[0056] Under the influence of wind, a target object in the sky will move due to force. In this case, the target object will appear in different positions in the sky images captured by the image acquisition device at different time points. By associating the sky images with the target object at multiple time points, the movement trajectory of the target object in the sky can be roughly obtained. This movement trajectory can serve as important data for subsequent wind field inversion. In order to improve the reliability of the inversion, as a preferred implementation method, in this scheme S02, when annotating the obtained sky image, the elements in the sky image that are not against a solid sky background are also annotated with outlines, and reference points are further marked on them according to the outline annotations of the elements. The number of reference points is at least one. When there is one, it is the geometric center of the outline annotation. When there is more than one, at least one reference point is located at the geometric center of the outline annotation, and the remaining reference points are located on the edge of the target object. The virtual connection between all adjacent reference points does not form a regular outline, that is, it does not form a circle, rectangle or other regular polygon.

[0057] In this scheme, the contour annotation of the sky image is in the form of two-dimensional contour data, which is used as intermediate data for pre-annotation. Subsequently, in step S05, when determining the location information of the target object in the same group of sky images based on at least three sky images and the location information of the associated image acquisition device, the obtained target object location information has a certain error. However, this error has a corresponding tolerance in estimating the target object location, meaning the error is acceptable. Regarding the problem of how to select the final reference point from multiple sky images, this scheme can either choose one method or fit the target positions from multiple sky images according to preset weights, for example:

[0058] X = n1×X1 + n2×X2 + ... + n m ×X m

[0059] n1 + n2 + ... + n1 = 1

[0060] Among them, X m After the location is determined, n represents the X-direction position (two-dimensional and three-dimensional reference coordinates are constructed according to a preset method) of different sky image reference points. m For X m The corresponding weight value is a positive number less than 1.

[0061] It should be noted that when there is only one reference point, its reference value for obtaining two-dimensional wind field information in S06 is better, because it is difficult to obtain the attitude change of the target object under the influence of the wind field with a single reference object. When two or more reference points are introduced, reference lines can be formed, and the attitude change of the target object can be better obtained by using reference lines. However, when there are two or more reference points, at least one reference point is located at the geometric center of the contour annotation, and the remaining reference points are located at the edge of the target object. Furthermore, the virtual connection between all adjacent reference points does not form a regular contour, that is, it does not form a circle, rectangle, or other regular polygon. This is because the shape of a regular contour may not change significantly when the target object rotates. For example, if a circular contour rotates around its center, the shape after rotation is almost the same as the shape before rotation, and it is not easy to change the floating state. In this case, more refined wind field information can be obtained by obtaining the attitude change of the reference points. For example, when the wind field acts on the target object at a specific angle, the situation of the target object rotating or moving in other non-unidirectional directions can be represented by multiple reference points.

[0062] It should be further explained that existing technologies support the research on inferring the actual position of a target object by combining three or more images with the location information of the shooting device, and then reflecting it in the established three-dimensional virtual coordinate system. For example, in the field of mechanical parts scanning, multi-angle photos of mechanical parts are used to reconstruct a three-dimensional model of the mechanical parts, and multi-angle images are used to obtain the actual position of distant objects. This solution cleverly applies this to the determination of the actual position of the target object in wind field measurement. Therefore, the computer theory will not be elaborated further. What needs to be understood is that the reference point strategy introduced in this solution to improve the trajectory recording of the target object is a unique mechanism of this solution.

[0063] The target objects described in this plan are those that float in the sky solely due to their own gravity and the force of the wind, including clouds, kites with loose strings, balloons with loose strings, and leaves.

[0064] In this scheme, all target objects mentioned are target objects with independent edge contours. They are not local segments of a whole object, because segmenting a whole object locally will increase the reference error. Taking cloud or contrail as an example, if the cloud or contrail is segmented to form multiple sub-target objects, the different forces on different areas of the same cloud will cause interference in trajectory judgment. For example, local turbulence in the atmosphere will act on the cloud locally, becoming an interfering factor. In addition, aircraft contrails or other slender target objects are prone to measurement interference due to their large span, which makes them susceptible to the influence of wind turbulence. Therefore, in S03 of this scheme, the edge of the target object is completely located in the sky image, and its proportion in the sky image is greater than 5% and less than 40%. In this solution, the target object is suitable as a cluster or block of clouds or a similar floating reference object. Regarding the selection method for the target object, as a preferred implementation method, in S03 of this solution, the method of determining the target object in a preset area of ​​the sky according to preset conditions includes: specifying the target object in a sky image that meets the preset conditions; when the specified target object has a contour annotation in the information annotation of the sky image, the target object is determined to meet the requirements, and then its features are extracted. Alternatively, a digital twin model can be built from the sky image captured by the image acquisition device, and the target object can be pre-selected by remotely selecting the target object in the digital twin model, combined with the aforementioned judgment rules to determine the target object.

[0065] Regarding trajectory fitting for the target object, a mathematical model can be fitted to the positions of the same reference point at different time points. This involves establishing a three-dimensional coordinate system reflecting the changes in the reference point's position. Then, based on the positional relationships of the reference point at different time points, a three-dimensional mathematical model is fitted to show the relationship between the reference point's positional changes and time. This allows for the prediction of the target object's position and state under the influence of wind. The estimated wind field information can then be inferred from this model. For example, after obtaining the three-dimensional mathematical model, a future time point is input to obtain the predicted position of the reference point. Then, after the future time point is reached, the distance difference between the actual and predicted positions is determined, thus revealing the reliability of the three-dimensional mathematical model and providing a rough assessment of whether the wind field is stable. In the case of an unstable wind field, wind speed and direction may change in real time, and the wind field information obtained from sky images has a limited reference value over time. When the target object has multiple reference points, fitting a three-dimensional mathematical model to the trajectory of each reference point can further provide a rough estimate of the wind direction acting on the target object under changing attitude conditions.

[0066] In this embodiment, the image capturing device can take various forms; it can be a specific image capturing device or an existing device solution with image capturing and data transmission functions, combined with... Figure 2 As shown, based on the above, this embodiment also provides the application of a handheld terminal as an image acquisition device in atmospheric wind field measurement, which includes the wind field measurement method based on the sky polarization background movement described above.

[0067] The handheld terminal is a mobile phone, handheld tablet computer, or camera equipped with a communication module, an image capture module, a pose sensor, and a positioning module.

[0068] In addition, a polarizing filter is provided on the lens of the image acquisition module.

[0069] Combination Figure 3 As shown, based on the above, the present invention also provides a wind field measurement system based on the movement of the sky polarization background, which includes:

[0070] The image acquisition devices are multiple and each is equipped with a polarizing filter. The multiple image acquisition devices capture images of a preset area of ​​the sky from different shooting angles to obtain multiple sky images.

[0071] Storage unit, used to store sky images captured by the image capturing device;

[0072] An information annotation unit is used to annotate the acquired sky images and then store them to generate a sky image set; wherein, the content of the information annotation includes at least the time point when the sky image was captured and the location information of the image capturing device that captured the sky image;

[0073] The target determination unit is used to determine the target object in a preset area of ​​the sky according to preset conditions.

[0074] The data matching unit is used to match the sky image set according to the characteristics of the target object to obtain multiple sky images with the target object. Then, the sky images with the same shooting time are grouped into a group to obtain at least two groups of sky images. Each group of sky images contains at least three sky images, and the corresponding image capturing devices are different.

[0075] The target positioning unit is used to determine the location information of a target object in the same set of sky images based on at least three sky images in the same set and the location information of the image acquisition device associated with them.

[0076] The data processing unit is used to invert the wind field in the area where the target object is located based on the shooting time point corresponding to at least two sets of sky images and the location information of the target object, thereby obtaining wind field information.

[0077] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0078] The storage unit, information labeling unit, target determination unit, data matching unit, target positioning unit, and data processing unit can all be integrated into a server for use. If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods of various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0079] The above description is only a part of the embodiments of the present invention and does not limit the scope of protection of the present invention. Any equivalent device or equivalent process transformation made based on the content of the present invention specification and drawings, or direct or indirect application in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for measuring wind field based on the movement of sky polarization background, characterized in that, include: S01. Multiple sky images are obtained by capturing images of a preset area of ​​the sky from different shooting angles using multiple image capturing devices equipped with polarizing filters. S02. The obtained sky images are labeled with information and then stored to generate a sky image set; wherein, the information labeling content includes at least the time point when the sky image was captured and the location information of the image capturing device that captured the sky image; S03. Determine the target object in the preset area of ​​the sky according to the preset conditions; S04. Match the sky images in the sky image set according to the characteristics of the target object to obtain multiple sky images with the target object. Then, group the sky images with the same shooting time into a group to obtain at least two groups of sky images. Each group of sky images contains at least three sky images, and the corresponding image capturing devices are different. S05. Determine the location information of the target object in the same set of sky images based on at least three sky images in the same set and the location information of the image capturing device associated with them; S06. Based on the shooting time points and the location information of the target object corresponding to at least two sets of sky images, the wind field in the area where the target object is located is inverted to obtain wind field information. Among the multiple image capturing devices, at least one image capturing device is included, and the image capturing range of one of the image capturing devices overlaps with that of the other image capturing device by 30% to 70%. In S02, before annotating the obtained sky image, the elements with non-sky solid color backgrounds in the images captured by two or more cameras of the image acquisition device are located and selected. Then, sky images with only sky solid color backgrounds and interfering elements are removed. The remaining sky images are then judged for clarity. Finally, the image with the highest element clarity is retained so that each element with a non-sky solid color background in the sky image of the same sky area corresponds to a sky image, generating at least one sky image. In S03, the edge of the target object is completely located in the sky image, and its occupancy ratio in the sky image is greater than 5% and less than 40%. The target objects are those that float in the sky solely due to their own gravity and the force of the wind, including clouds, kites with loose strings, balloons with loose strings, and leaves.

2. The wind field measurement method based on sky polarization background movement as described in claim 1, characterized in that, In S01, each image capturing device has two or more cameras, and the two or more cameras take pictures simultaneously with different focusing parameters. The captured photos are processed according to preset conditions to generate sky images. Among them, the image capture range of multiple image capturing devices completely covers the preset area of ​​the sky.

3. The wind field measurement method based on sky polarization background movement as described in claim 2, characterized in that, In S01, two or more cameras of the image acquisition device automatically focus on a preset area of ​​the sky using a multi-point focusing method to obtain focus parameters and take pictures. When the focus parameters of two or more cameras are the same, the focusing is re-executed so that the focus parameters of different cameras are not exactly the same. The two or more cameras take pictures synchronously with not exactly the same focus parameters to obtain multiple sky images that meet the preset requirements and upload them to the server.

4. The wind field measurement method based on sky polarization background movement as described in claim 3, characterized in that, In S02, when annotating the obtained sky image, the elements in the sky image that are not in a solid sky background are also annotated with contours, and reference points are further marked on the elements according to their contour annotations. The number of reference points is at least one. When there is one reference point, it is the geometric center of the contour annotation. When there is more than one reference point, at least one reference point is located at the geometric center of the contour annotation.

5. The wind field measurement method based on sky polarization background movement as described in claim 1, characterized in that, In S03, the method of determining the target object in the preset area of ​​the sky according to preset conditions includes: specifying the target object in the sky image that meets the preset conditions; when the specified target object has a contour annotation in the information annotation of the sky image, it is determined that the target object meets the requirements, and then its features are extracted.

6. The wind field measurement method based on sky polarization background movement as described in claim 5, characterized in that, In S01, the polarizing mirror mounted on the image capturing device is used to filter stray light in the shooting environment.

7. The application of a handheld terminal as an image acquisition device in atmospheric wind field measurement, characterized in that, It includes the wind field measurement method based on the movement of the sky polarization background as described in any one of claims 1 to 6; The handheld terminal is a mobile phone, handheld tablet computer, or camera equipped with a communication module, an image capture module, a pose sensor, and a positioning module. In addition, a polarizing filter is provided on the lens of the image acquisition module.

8. A wind field measurement system based on sky polarization background movement, which applies the wind field measurement method based on sky polarization background movement as described in any one of claims 1 to 6, characterized in that, It includes: The image acquisition devices are multiple and each is equipped with a polarizing filter. The multiple image acquisition devices capture images of a preset area of ​​the sky from different shooting angles to obtain multiple sky images. Storage unit, used to store sky images captured by the image capturing device; An information annotation unit is used to annotate the acquired sky images and then store them to generate a sky image set; wherein, the content of the information annotation includes at least the time point when the sky image was captured and the location information of the image capturing device that captured the sky image; The target determination unit is used to determine the target object in a preset area of ​​the sky according to preset conditions. The data matching unit is used to match the sky image set according to the characteristics of the target object to obtain multiple sky images with the target object. Then, the sky images with the same shooting time are grouped into a group to obtain at least two groups of sky images. Each group of sky images contains at least three sky images, and the corresponding image capturing devices are different. The target positioning unit is used to determine the location information of a target object in the same set of sky images based on at least three sky images in the same set and the location information of the image acquisition device associated with them. The data processing unit is used to invert the wind field in the area where the target object is located based on the shooting time point corresponding to at least two sets of sky images and the location information of the target object, thereby obtaining wind field information.

9. A computer-readable storage medium, characterized in that: The storage medium stores at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, at least one program, code set, or instruction set is loaded and executed by a processor to implement the wind field measurement method based on the movement of the sky polarization background as described in any one of claims 1 to 6.

Images