Cloud position and cloud velocity measurement method and device based on all-sky cloud map network

Abstract

The present invention relates to a cloud position and cloud speed measurement method, comprising the following steps: obtaining the full sky image taken by each full sky imager and the coordinate point relationship between the pixel coordinates and the spatial coordinates in the three-dimensional spatial coordinate system; identifying the spatial feature matching points of the full sky images taken by the paired full sky imagers at the same time; calculating the azimuth of the connecting line between the two points in each spatial feature point pair; calculating the elevation and azimuth of the cloud feature point; calculating the spatial coordinates of the cloud feature point; calculating the correlation between the pixel coordinate change and the spatial coordinate change of the same full sky imager at two adjacent moments; and calculating the cloud movement speed. In the technical solution of the present invention, the network observation of ground-based cloud maps is utilized, and on the basis of the three-dimensional spatial positioning of the feature points on the cloud cluster in the cloud map, the time series of the full sky cloud map is analyzed and modeled to achieve quantitative observation of the cloud movement speed, thereby providing more accurate data support for cloud movement prediction and new energy applications.

Description

Technical Field

[0001] The present invention relates to the field of meteorological data processing technology, and in particular to a cloud position and cloud speed measurement method, system, readable storage medium and computer equipment based on all-sky cloud map networking. Background Art

[0002] Cloud obstruction is a significant factor affecting the efficiency of photovoltaic power generation. The difficulty and key to photovoltaic power forecasting lies in predicting the local cloud movement speed and direction. Cloud forecasting often requires the current true or observed cloud movement speed as input to obtain a relatively accurate local cloud movement speed forecast. However, due to limitations in observation methods, current cloud movement speed observation data is extremely scarce, which is a major factor limiting the accuracy of cloud movement forecasts.

[0003] The means of observing clouds include three methods: spaceborne remote sensing, ground-based radar remote sensing and ground-based all-sky imagers.

[0004] The most commonly used method for obtaining large-scale, large-scale cloud movement speed and direction observation data is to calculate cloud-guided winds using satellite remote sensing images. However, the spatial resolution of satellite observation methods is often above 1 km, which is a coarse resolution, and the local observation time is often greater than 1 hour. This cannot meet the requirements of a specific individual photovoltaic power station for short-term photovoltaic power forecasting, especially the forecast needs in the near future.

[0005] Ground-based radar remote sensing observations primarily use millimeter-wave cloud radar or lidar to scan the sky from bottom to top. This observation method actively transmits millimeter waves or lasers and measures the signal return time and phase information to obtain cloud height and layer information. It can also estimate cloud cover information based on cloud motion and obstruction. However, this type of equipment has significant drawbacks. First, the scanning angle is often very small, making it difficult to quickly obtain full-sky cloud cover information. Second, this type of equipment cannot determine the horizontal movement speed of clouds, which are often quasi-horizontal. Finally, the acquisition and maintenance costs of active remote sensing radar equipment are often high, limiting the methods used.

[0006] Ground-based all-sky imagers typically observe the entire sky using visible-light cameras using a fisheye projection method. Using projection transformation and pixel extraction of cloud-covered areas, they can obtain cloud distribution within a radius of approximately 2 kilometers. This high spatial and temporal resolution allows for accurate estimation of cloud movement through time-series analysis. However, images from a single ground-based all-sky imager only provide azimuth and angular information, preventing observation of cloud height and distance, and thus, accurate quantitative measurement of cloud movement speed. Networked observations can provide cloud observation information from different angles, enabling further calculation of cloud height and spatial coordinates. For example, Tao et al. (2013) proposed a method for measuring cloud base height based on binocular imaging. This method established an observation system with a baseline length of 60 meters and obtained relatively accurate cloud base height from two simultaneous images. However, this system did not address feature matching between different time frames and assumed that the two cameras were at the same altitude, a requirement that is difficult to achieve in a system with two all-sky imagers at a distance. Similarly, Li Guosheng (2011) systematically discussed a method and system for binocular cloud base height measurement in his doctoral dissertation. However, this system used a cloud camera rather than an all-sky imager. Regarding the method of using dual cameras for cloud motion observation, the patent "A Method, System, and Process for Cloud Tracking and Prediction in Photovoltaic Power Stations, Application No. CN202110850460.3" proposes, from a process and conceptual perspective, using two cameras to capture two real-time videos of the area to be tracked; decomposing the two real-time videos into two corresponding sets of frame images, and using a trained object detection and segmentation model to identify the sun and multiple cloud regions in the two sets of frame images; and calculating the cloud height, cloud velocity, and cloud range for each of the cloud regions. However, the patent does not provide specific calculation methods and implementation steps, only providing a general execution process. The patent also contains significant ambiguity, providing only module information, rather than the calculation method, for key information such as camera projection method, installation, cloud feature matching method, and cloud height calculation method. Therefore, the patent's guidance and specificity are limited. Summary of the Invention

[0007] The present invention aims to solve at least one of the technical problems existing in the prior art or related art.

[0008] To this end, one object of the present invention is to provide a cloud position and cloud speed measurement method based on the all-sky cloud map network, which utilizes ground-based cloud map network observations, and on the basis of the three-dimensional spatial positioning of characteristic points on the cloud clusters in the cloud map, analyzes and models the all-sky cloud map time series to achieve quantitative observation of the cloud movement speed, providing more accurate data support for cloud movement prediction and new energy applications.

[0009] To achieve the above objectives, the technical solution of the first aspect of the present invention provides a cloud position and cloud velocity measurement method based on a full sky cloud map network, comprising the following steps:

[0010] Obtain the full sky image taken by each full sky imager and the relationship between the pixel coordinates of the corresponding full sky image and the coordinate points of the spatial coordinates in the three-dimensional spatial coordinate system, specifically:

[0011]

[0012] Wherein, the field of view (FOV) of the all-sky imager is 180°, N is the number of pixels in the all-sky image, and the point at the i-th pixel in the horizontal direction and the j-th pixel in the vertical direction of the all-sky image corresponds to the spatial coordinates (x, y, z) of the point P in the three-dimensional space coordinate system;

[0013] Identify the spatial feature matching points of the all-sky images taken by the paired all-sky imagers at the same time to obtain spatial feature point pairs;

[0014] Calculate the direction angle of the line connecting the two points in each spatial feature point pair, specifically:

[0015] Among them, the two full sky images I at time t 1，t and I 2，t The coordinates of P are 1，t，k (i 1，t，k ,j 1，t，k ) and P 2，t，k (i 2，t，k ,j 2，t，k );

[0016] The spatial feature point pairs are quality controlled according to the preset threshold corresponding to the direction angle to obtain the matching feature point set. The expression of the matching feature point set PF is:

[0017] P F =[(P 1，t，k , P 2，t，k ) if θ t，k <median(θ t )-10 |k=1,2,...,N];where, θ t，k Represented as the kth matching spatial feature point pair P at time t 1，t，k (i 1，t，k ,j 1，t，k ) and P 2，t，k (i 2，t，k ,j 2，t，k )’s connecting line direction angle, median(θ t ) is expressed as the average value of the direction angles of the connecting lines of all spatial feature point pairs at time t. The preset threshold is that the difference between the direction angle of the connecting line of the kth matching spatial feature point pair at time t and the average value of the direction angles of the connecting lines of all spatial feature point pairs at time t is less than 10, that is, θ t，k<median(θ t )-10;

[0018] Calculate the elevation and azimuth angles of cloud feature points based on the pixel coordinates of the matching feature point set in the full sky image;

[0019] Calculate the spatial coordinates of the cloud feature points according to the elevation angle and azimuth angle of the cloud feature points and the spatial coordinates of the paired all-sky imagers in the three-dimensional spatial coordinate system;

[0020] The correlation between the pixel coordinate change and the spatial coordinate change of the same all-sky imager at two adjacent moments is calculated based on the spatial coordinates of the cloud feature points;

[0021] The cloud movement speed is calculated based on the correlation between the pixel coordinate change and the spatial coordinate change.

[0022] In the above technical solution, preferably, calculating the elevation angle and azimuth angle of the cloud feature point according to the pixel coordinates of the matching feature point set in the full sky image includes the following steps:

[0023] The azimuth angle β and elevation angle α of each all-sky imager are calculated based on the spatial coordinates of the all-sky imager and the relationship between the pixel coordinates and the spatial coordinates. Specifically,

[0024]

[0025] Substitute the pixel coordinates of the cloud feature point in the matching feature point set into the azimuth and elevation angles of the all-sky imager to obtain the elevation angle α of the cloud feature point 1，t，k and α 2，t，k and the azimuth angle β 1，t，k and β 2，t，k .

[0026] In the above technical solution, preferably, calculating the spatial coordinates of the cloud feature points according to the elevation angle and azimuth angle of the cloud feature points and the spatial coordinates of the paired all-sky imagers in a three-dimensional spatial coordinate system includes the following steps:

[0027] According to the definition of trigonometric tangent function, the positional relationship between cloud feature points and paired all-sky imagers in the three-dimensional space coordinate system is obtained as follows:

[0028]

[0029] Among them, the spatial coordinates of the paired sky imagers are A1(x1, y1, z1) and A2(x2, y2, z2), and the cloud feature point P c The spatial coordinates (x c ,y c , z c ), the elevation angles α1 and α2 of the paired sky imagers;

[0030] According to the cosine theorem of triangles, the distance relationship between cloud feature points and pairs of all-sky imagers in the three-dimensional space coordinate system is obtained, and the expression is:

[0031] d xy，12 2 =d xy，1c 2 +d xy，2c 2 -2×d xy，1c ×d xy，2c ×cos(|β1-β2|); where d xy，12 Refers to the horizontal distance between two all-sky imagers, d xy，1c and d xy，2c They refer to the horizontal distances between the two all-sky imagers and the cloud feature points;

[0032]

[0033] The spatial coordinates of the cloud feature points in the three-dimensional space coordinate system are calculated based on the elevation and azimuth angles, positional relationships, and distance relationships of the cloud feature points. The expression is:

[0034] in,

[0035]

[0036] Then, according to the coordinate point relationship between pixel coordinates and spatial coordinates, we get:

[0037]

[0038] The elevation angle α 1，t，k and α 2，t，k , and the azimuth angle β 1，t，k and β 2，t，k Substituting the above formula, we can get the spatial coordinate z of the cloud feature point in the three-dimensional space coordinate system: c，t，k , then z c，t，k Substituting the coordinate point relationship between pixel coordinates and spatial coordinates, we can get the spatial coordinates (x c，t，k ,y c，t，k ).

[0039] In the above technical solution, preferably, the pixel coordinate change is:

[0040] Among them, the pixel coordinates of the spatial feature points captured by any all-sky imager at two adjacent moments are (i1, j1) and (i2, j2); similarly, the kth matching spatial feature point pair P is obtained. 1，t，k (i 1，t，k ,j1，t，k ) and P 2，t，k (i 2，t，k ,j 2，t，k ) pixel coordinate change Δi t，k and Δj t，k ;

[0041] Then, according to the coordinate point relationship between pixel coordinates and spatial coordinates, the spatial coordinate change is obtained as:

[0042]

[0043] will i 1，t，k and j 1，t，k , and Δi t，k and Δj t，k and z c，t，k Substituting into the above formula, we can get the spatial coordinate change Δx c，t，k and Δy c，t，k ;

[0044] The cloud movement speed is calculated based on the correlation between the pixel coordinate change and the spatial coordinate change, specifically:

[0045] Where Δt is the time interval between two adjacent moments, so we can get the spatial coordinates of the kth feature point and the cloud moving speed at time t, that is, V c，t，k =(x c，t，k ,y c，t，k , z c，t，k ,u c，t，k , v c，t，k ), all observation data sets at time t are V c，t =(V c，t，1 , V c，t，2 ,…,V c，t，M ), where M is the number of valid cloud feature points.

[0046] The technical solution of the second aspect of the present invention provides a cloud position and cloud velocity measurement system based on a full sky cloud map network, comprising:

[0047] The pixel coordinate and spatial coordinate conversion module is configured to obtain the full sky image captured by each full sky imager and the coordinate point relationship between the pixel coordinates of the corresponding full sky image and the spatial coordinates in the three-dimensional spatial coordinate system, specifically:

[0048]

[0049] Wherein, the field of view (FOV) of the all-sky imager is 180°, N is the number of pixels in the all-sky image, and the point at the i-th pixel in the horizontal direction and the j-th pixel in the vertical direction of the all-sky image corresponds to the spatial coordinates (x, y, z) of the point P in the three-dimensional space coordinate system;

[0050] A spatial feature recognition module is configured to identify spatial feature matching points of all-sky images captured by a pair of all-sky imagers at the same time, thereby obtaining spatial feature point pairs;

[0051] The connection direction angle calculation module is configured to calculate the direction angle of the connection line between two points in each spatial feature point pair, specifically:

[0052] Among them, the two full sky images I at time t 1，t and I 2，t The coordinates of P are 1，t，k (i 1，t，k ,j 1，t，k ) and P 2，t，k (i 2，t，k ,j 2，t，k );

[0053] The data quality control module is configured to perform quality control on the spatial feature point pairs according to a preset threshold corresponding to the direction angle, and obtain a matching feature point set. The matching feature point set P F The expression is:

[0054] P F =[(P 1，t，k , P 2，t，k ) if θ t，k <median(θ t )-10 |k=1,2,...,N];where, θ t，k Represented as the kth matching spatial feature point pair P at time t 1，t，k (i 1，t，k ,j 1，t，k ) and P 2，t，k (i 2，t，k ,j 2，t，k )’s connecting line direction angle, median(θ t ) is expressed as the average value of the direction angles of the connecting lines of all spatial feature point pairs at time t. The preset threshold is that the difference between the direction angle of the connecting line of the kth matching spatial feature point pair at time t and the average value of the direction angles of the connecting lines of all spatial feature point pairs at time t is less than 10, that is, θ t，k <median(θ t )-10;

[0055] a cloud feature angle calculation module configured to calculate the elevation angle and azimuth angle of a cloud feature point based on the pixel coordinates of the matching feature point set in the full sky image;

[0056] a cloud feature point spatial calculation module configured to calculate the spatial coordinates of the cloud feature points based on the elevation angle and azimuth angle of the cloud feature points and the spatial coordinates of the paired all-sky imagers in a three-dimensional spatial coordinate system;

[0057] a cloud feature coordinate change conversion module configured to calculate, based on the spatial coordinates of the cloud feature points, a correlation between a pixel coordinate change and a spatial coordinate change of the same all-sky imager at two adjacent moments;

[0058] The cloud speed calculation module is configured to calculate the cloud movement speed based on the correlation between the pixel coordinate change amount and the space coordinate change amount.

[0059] In the above technical solution, preferably, the cloud feature angle calculation module includes:

[0060] The imager angle calculation unit is configured to calculate the azimuth angle β and elevation angle α of each all-sky imager based on the spatial coordinates of the all-sky imager and the coordinate point relationship between the pixel coordinates and the spatial coordinates, specifically:

[0061]

[0062] The cloud feature angle calculation unit is configured to substitute the pixel coordinates of the cloud feature point in the matching feature point set into the azimuth and elevation angles of the all-sky imager to obtain the elevation angle α of the cloud feature point. 1，t，k and α 2，t，k and the azimuth angle β 1，t，k and β 2，t，k .

[0063] In the above technical solution, preferably, the cloud feature point space calculation module includes:

[0064] The position relationship calculation unit is configured to obtain the position relationship between the cloud feature point and the paired all-sky imager in the three-dimensional space coordinate system according to the definition of the trigonometric tangent function, and the expression is:

[0065]

[0066] Among them, the spatial coordinates of the paired sky imagers are A1(x1, y1, z1) and A2(x2, y2, z2), and the cloud feature point P c The spatial coordinates (x c ,y c , z c ), the elevation angles α1 and α2 of the paired sky imagers;

[0067] The distance relationship calculation unit is configured to obtain the distance relationship between the cloud feature points and the all-sky imagers in a three-dimensional space coordinate system according to the cosine theorem of a triangle. The expression is:

[0068] d xy，12 2 =d xy，1c 2 +d xy，2c 2 -2×d xy，1c ×d xy，2c ×cos(|β1-β2|); where d xy，12 Refers to the horizontal distance between two all-sky imagers, d xy，1c and d xy，2c They refer to the horizontal distances between the two all-sky imagers and the cloud feature points;

[0069]

[0070]

[0071] The cloud feature point spatial calculation unit is configured to calculate the spatial coordinates of the cloud feature points in the three-dimensional space coordinate system based on the elevation angle and azimuth angle, position relationship and distance relationship of the cloud feature points. The expression is:

[0072] in,

[0073]

[0074] Then, according to the coordinate point relationship between pixel coordinates and spatial coordinates, we get:

[0075]

[0076] Finally, the elevation angle α 1，t，k and α 2，t，k , and the azimuth angle β 1，t，k and β 2，t，k Substituting the above formula, we can get the spatial coordinate z of the cloud feature point in the three-dimensional space coordinate system: c，t，k , then z c，t，k Substituting the coordinate point relationship between pixel coordinates and spatial coordinates, we can get the spatial coordinates (x c，t，k ,y c，t，k ).

[0077] In the above technical solution, preferably, the pixel coordinate change is:

[0078] Among them, the pixel coordinates of the spatial feature points captured by any all-sky imager at two adjacent moments are (i1, j1) and (i2, j2); similarly, the kth matching spatial feature point pair P is obtained. 1，t，k (i 1，t，k ,j 1，t，k ) and P 2，t，k (i 2，t，k ,j 2，t，k ) pixel coordinate change Δi t，k and Δj t，k ;

[0079] The change in spatial coordinates is:

[0080]

[0081] i 1，t，k and j 1，t，k , and Δi t，k and Δj t，k and z c，t，k Substituting into the above formula, we can get the spatial coordinate change Δx c，t，k and Δy c，t，k ;

[0082] The expression for cloud movement speed is:

[0083] Where Δt is the time interval between two adjacent moments, so we can get the spatial coordinates of the kth feature point and the cloud moving speed at time t, that is, V c，t，k =(x c，t，k ,y c，t，k , z c，t，k ,u c，t，k , v c，t，k ), all observation data sets at time t are V c，t =(V c，t，1 , V c，t，2 ,…,V c，t，M ), where M is the number of valid cloud feature points.

[0084] The technical solution of the third aspect of the present invention provides a readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps of the cloud position and cloud speed measurement method based on the all-sky cloud map network provided by the technical solution of the first aspect are implemented.

[0085] The technical solution of the fourth aspect of the present invention provides a computer device, including a storage medium and a processor; the storage medium is used to store a computer program; the processor is used to execute the computer program to implement the steps of the cloud position and cloud speed measurement method based on the all-sky cloud map network provided by the technical solution of the first aspect above.

[0086] Compared with the existing technology, the advantages of the cloud position and cloud speed measurement method and equipment based on the all-sky cloud map network provided by the present invention are: utilizing ground-based cloud map network observation, on the basis of the three-dimensional spatial positioning of characteristic points on the cloud clusters in the cloud map, the all-sky cloud map time series is analyzed and modeled, to achieve quantitative observation of the cloud movement speed, and provide more accurate data support for cloud movement prediction and new energy applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments with reference to the accompanying drawings, in which:

[0088] Figure 1 A schematic diagram of a three-dimensional spatial coordinate system centered on the all-sky imager is shown;

[0089] Figure 2 A schematic diagram of the imaging principle of the all-sky imager is shown;

[0090] Figure 3 Schematic diagram showing the connecting lines of the original and matched Shi-Tomasi feature points;

[0091] Figure 4 A schematic diagram of a bottom-height intelligent observation method is shown;

[0092] Figure 5 A flowchart of a method according to an embodiment of the present invention is shown;

[0093] Figure 6 shows a flowchart of step S5 involved in an embodiment of the present invention;

[0094] Figure 7 shows a flowchart of step S6 involved in an embodiment of the present invention;

[0095] Figure 8 shows a structural block diagram of a system involved in an embodiment of the present invention;

[0096] Figure 9 The following is a structural block diagram of a cloud feature angle calculation module according to an embodiment of the present invention;

[0097] Figure 10 The figure shows a structural block diagram of a cloud feature point space calculation module involved in an embodiment of the present invention. DETAILED DESCRIPTION

[0098] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present application and the features therein can be combined with each other.

[0099] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0100] like Figures 1 to 7 As shown, according to an embodiment of the present invention, a cloud position and cloud velocity measurement method based on a full sky cloud map network includes the following steps:

[0101] S1, obtaining the full sky image captured by each full sky imager and the relationship between the pixel coordinates of the corresponding full sky image and the coordinate points of the spatial coordinates in the three-dimensional spatial coordinate system, specifically:

[0102] Assume that the center of the image is the origin, and assume that we have a point in the image, which is the i-th pixel in the horizontal direction and the j-th pixel in the vertical direction. At the same time, the actual position of the object corresponding to this pixel in the three-dimensional coordinate system of the space world is P(x, y, z). Then according to the relationship of isometric projection ( Figure 2 ), we get the relationship between the image coordinates and the world coordinate system of the pixels in the image as follows:

[0103]

[0104] Wherein, the field of view (FOV) of the all-sky imager is 180°, N is the number of pixels in the all-sky image, and the point at the i-th pixel in the horizontal direction and the j-th pixel in the vertical direction of the all-sky image corresponds to the spatial coordinates (x, y, z) of the point P in the three-dimensional space coordinate system;

[0105] In this step, the present invention uses two or more all-sky imagers for networked observation. Since the method of calculating cloud base height and cloud movement speed by matching two all-sky cloud images is adopted, if there are more than two all-sky imagers, cloud height and cloud movement speed can be calculated by pairing them one by one. Figure 1 As shown, this system features two all-sky imagers, A1 and A2, installed at positions A1(x1, y1, z1) and A1(x2, y2, z2), respectively. Each all-sky imager is installed in a standardized manner, meeting the following two conditions: 1) The imager is installed vertically, with the center of the all-sky cloud image at the imager's zenith. 2) The imager's azimuth angle ensures that the column direction of the captured image corresponds to the actual north-south direction, and the row direction corresponds to the actual east-west direction.

[0106] After the installation of the all-sky imager is completed, the all-sky imager is synchronized and calibrated to ensure that the time of the two cameras is consistent. When actually shooting images, the two all-sky imagers shoot at the same time. The all-sky images obtained by A1 and A2 at time t are I 1，t and I 2，t In order to record and analyze the movement of cloud images, two all-sky imagers are controlled to take synchronous and continuous images of the sky. The time interval between the two all-sky imagers is Δt.

[0107] In this network observation scheme, the projection method of the image taken by each all-sky imager is equidistant projection, such as Figure 2 Under the condition of equidistant projection, assuming that the diameter of the full sky cloud map is N, and the FOV of the full sky imager is 180°, then a single pixel represents a consistent viewing angle and

[0108] S2, identifying the spatial feature matching points of the all-sky images taken by the paired all-sky imagers at the same time, and obtaining spatial feature point pairs;

[0109] S3, calculate the direction angle of the line connecting the two points in each spatial feature point pair, specifically:

[0110] Among them, the two full sky images I at time t 1，t and I 2，t The coordinates of P are 1，t，k (i 1，t，k ,j 1，t，k ) and P 2，t，k (i 2，t，k ,j 2，t，k );

[0111] In this step, in order to calculate the cloud base height, we need to match the cloud features in the paired full-sky images. In this study, we implemented the Shi-Tomasi-SIFT corner detection and matching algorithm to obtain and match cloud features. The algorithm uses the Shi-Tomasi algorithm to detect candidate good features, and then uses the SIFT algorithm to describe and match these features in the two images of the paired camera. The original image is normalized to 0-255 to minimize the impact of brightness changes. The Shi-Tomasi algorithm is insensitive to viewing angle, brightness, rotation, etc., and is computationally efficient. The SIFT algorithm is very effective in describing the direction and scale of features, and is also very robust to brightness, scale and rotation. The combination of Shi-Tomasi and SIFT algorithms can accurately and quickly detect the features of cloud images. According to this method, assuming that in a set of images I from two stations 1，t and I 2，tAfter corner point detection and matching, a set of N matching points is obtained. For the k-th spatial feature point matched, the two images I at time t 1，t and I 2，t The coordinates in are P 1，t，k (i 1，t，k ,j 1，t，k ) and P 2，t，k (i 2，t，k ,j 2，t，k ).

[0112] In order to evaluate the direction between feature points, the average direction of the connecting line between the matching feature points is calculated and the images of the two cameras at the same time are spliced ​​left and right, such as Figure 3 As shown, the left and right figures are I 1，t and I 2，t Considering that the all-sky cameras have the same installation azimuth, after the two pre-processed images are spliced ​​in left-right arrangement, the connecting lines of all feature points should be basically parallel and there will be no intersection at too large an angle, such as Figure 3 For each individual matching point pair, for example, at time t, the image coordinates of the kth matching feature point in the two images are P 1，t，k and P 2，t，k ,according to Figure 3 The arrangement shown describes the direction angle of the connecting line as θ, where the angle θ is 0° to the right of the horizontal direction and increases clockwise, and the kth connecting line direction angle θ can be obtained.

[0113] S4, quality control is performed on the spatial feature point pairs according to the preset threshold corresponding to the direction angle to obtain the matching feature point set, the matching feature point set P F The expression is:

[0114] P F =[(P 1，t，k , P 2，t，k ) if θ t，k <median(θ t )-10 |k=1,2,...,N];where, θ t，k Represented as the kth matching spatial feature point pair P at time t 1，t，k (i 1，t，k ,j 1，t，k ) and P 2，t，k (i 2，t，k ,j 2，t，k )’s connecting line direction angle, median(θ t) is expressed as the average value of the direction angles of the connecting lines of all spatial feature point pairs at time t. The preset threshold is that the difference between the direction angle of the connecting line of the kth matching spatial feature point pair at time t and the average value of the direction angles of the connecting lines of all spatial feature point pairs at time t is less than 10, that is, θ t，k <median(θ t )-10; the quality-controlled matching feature point set P F The number is M.

[0115] S5, calculating the elevation and azimuth angles of the cloud feature points based on the pixel coordinates of the matched feature point set in the full sky image;

[0116] S6, calculating the spatial coordinates of the cloud feature points according to the elevation angle and azimuth angle of the cloud feature points and the spatial coordinates of the paired all-sky imagers in the three-dimensional spatial coordinate system;

[0117] S7, calculating the correlation between the pixel coordinate change and the spatial coordinate change of the same all-sky imager at two adjacent moments according to the spatial coordinates of the cloud feature points;

[0118] S8, calculating the cloud movement speed based on the correlation between the pixel coordinate change and the space coordinate change.

[0119] like Figure 6 As shown, in the above embodiment, preferably, S5, calculating the elevation angle and azimuth angle of the cloud feature point according to the pixel coordinates of the matching feature point set in the full sky image, includes the following steps:

[0120] S51, calculating the azimuth angle β and elevation angle α of each all-sky imager based on the spatial coordinates of the all-sky imager and the coordinate point relationship between the pixel coordinates and the spatial coordinates, specifically:

[0121]

[0122] In this step, the feature set P obtained after quality control F It will be used to estimate the cloud movement speed. After finding the cloud feature points in the two images, the cloud height can be calculated based on the azimuth and elevation angles of the feature points in the two cloud images. The calculation process is shown in the figure below. Figure 4 As shown. In the network observation, the spatial coordinates of the two all-sky imagers A1(x1, y1, z1) and A1(x2, y2, z2) are known. At the same time, according to the imaging relationship of the all-sky imager Figure 2 , the azimuth angle β and elevation angle α can be calculated.

[0123] S52, substituting the pixel coordinates of the cloud feature point in the matching feature point set into the azimuth and elevation angles of the all-sky imager to obtain the elevation angle α of the cloud feature point 1，t，k and α 2，t，k and the azimuth angle β1，t，k and β 2，t，k .

[0124] like Figure 7 As shown, in the above embodiment, preferably, S6, calculating the spatial coordinates of the cloud feature points according to the elevation angle and azimuth angle of the cloud feature points and the spatial coordinates of the paired all-sky imagers in the three-dimensional spatial coordinate system, includes the following steps:

[0125] S61, according to the definition of the trigonometric tangent function, obtain the positional relationship between the cloud feature points and the paired all-sky imagers in the three-dimensional space coordinate system, expressed as:

[0126]

[0127] Among them, the spatial coordinates of the paired sky imagers are A1(x1, y1, z1) and A2(x2, y2, z2), and the cloud feature point P c The spatial coordinates (x c ,y c , z c ), the elevation angles α1 and α2 of the paired sky imagers;

[0128] S62, according to the cosine theorem of triangles, the distance relationship between the cloud feature points and the all-sky imagers in the three-dimensional space coordinate system is obtained, which is expressed as:

[0129] d xy，12 2 =d xy，1c 2 +d xy，2c 2 -2×d xy，1c ×d xy，2c ×cos(|β1-β2|); where d xy，12 Refers to the horizontal distance between two all-sky imagers, d xy，1c and d xy，2c They refer to the horizontal distances between the two all-sky imagers and the cloud feature points;

[0130]

[0131] S63, calculating the spatial coordinates of the cloud feature points in the three-dimensional space coordinate system based on the elevation angle and azimuth angle, position relationship, and distance relationship of the cloud feature points. The expression is:

[0132] in,

[0133]

[0134] Then, according to the coordinate point relationship between pixel coordinates and spatial coordinates, we get:

[0135]

[0136] The elevation angle α 1，t，k and α 2，t，k , and the azimuth angle β 1，t，k and β 2，t，k Substituting the above formula, we can get the spatial coordinate z of the cloud feature point in the three-dimensional space coordinate system: c，t，k , then z c，t，k Substituting the coordinate point relationship between pixel coordinates and spatial coordinates, we can get the spatial coordinates (x c，t，k ,y c，t，k ).

[0137] In the above embodiment, preferably, the pixel coordinate change is:

[0138] Among them, the pixel coordinates of the spatial feature points captured by any all-sky imager at two adjacent moments are (i1, j1) and (i2, j2); similarly, the kth matching spatial feature point pair P is obtained. 1，t，k (i 1，t，k ,j 1，t，k ) and P 2，t，k (i 2，t，k ,j 2，t，k ) pixel coordinate change Δi t，k and Δj t，k ;

[0139] In order to calculate the coordinate changes of the spatial points corresponding to the changes of the coordinate points in the cloud map, considering that the horizontal speed of the cloud is much greater than the vertical speed in most cases, it can be assumed that the height of the cloud remains basically unchanged and the cloud only moves in the horizontal direction.

[0140] Then, according to the coordinate point relationship between pixel coordinates and spatial coordinates, the spatial coordinate change is obtained as:

[0141]

[0142] will i 1，t，k and j 1，t，k , and Δi t，k and Δj t，k and z c，t，k Substituting into the above formula, we can get the spatial coordinate change Δx c，t，k and Δy c，t，k ;

[0143] The cloud movement speed is calculated based on the correlation between the pixel coordinate change and the spatial coordinate change, specifically:

[0144] Where Δt is the time interval between two adjacent moments, so we can get the spatial coordinates of the kth feature point and the cloud moving speed at time t, that is, V c，t，k =(x c，t，k ,y c，t，k , z c，t，k ,u c，t，k , v c，t，k ), all observation data sets at time t are V c，t =(V c，t，1 , V c，t，2 ,…,V c，t，M ), where M is the number of valid cloud feature points.

[0145] like Figures 1 to 4 、 Figures 8 to 10 As shown, according to another embodiment of the present invention, a cloud position and cloud velocity measurement system 100 based on a full sky cloud map network includes:

[0146] The pixel coordinate and spatial coordinate conversion module 1 is configured to obtain the full sky image captured by each full sky imager and the coordinate point relationship between the pixel coordinates of the corresponding full sky image and the spatial coordinates in the three-dimensional spatial coordinate system, specifically:

[0147]

[0148] Wherein, the field of view (FOV) of the all-sky imager is 180°, N is the number of pixels in the all-sky image, and the point at the i-th pixel in the horizontal direction and the j-th pixel in the vertical direction of the all-sky image corresponds to the spatial coordinates (x, y, z) of the point P in the three-dimensional space coordinate system;

[0149] The spatial feature recognition module 2 is configured to recognize spatial feature matching points of the full-sky images captured by the paired full-sky imagers at the same time, and obtain spatial feature point pairs;

[0150] The connection direction angle calculation module 3 is configured to calculate the direction angle of the connection line between two points in each spatial feature point pair, specifically:

[0151] Among them, the two full sky images I at time t 1，t and I 2，t The coordinates of P are 1，t，k (i 1，t，k ,j 1，t，k ) and P 2，t，k (i 2，t，k ,j 2，t，k )

[0152] The data quality control module 4 is configured to perform quality control on the spatial feature point pairs according to a preset threshold value corresponding to the direction angle, and obtain a matching feature point set. The matching feature point set PF The expression is:

[0153] P F =[(P 1，t，k , P 2，t，k ) if θ t，k <median(θ t )-10 |k=1,2,...,N];where, θ t，k Represented as the kth matching spatial feature point pair P at time t 1，t，k (i 1，t，k ,j 1，t，k ) and P 2，t，k (i 2，t，k ,j 2，t，k )’s connecting line direction angle, median(θ t ) is expressed as the average value of the direction angles of the connecting lines of all spatial feature point pairs at time t. The preset threshold is that the difference between the direction angle of the connecting line of the kth matching spatial feature point pair at time t and the average value of the direction angles of the connecting lines of all spatial feature point pairs at time t is less than 10, that is, θ t，k <median(θ t )-10;

[0154] a cloud feature angle calculation module 5, configured to calculate the elevation angle and azimuth angle of the cloud feature point based on the pixel coordinates of the matching feature point set in the full sky image;

[0155] a cloud feature point spatial calculation module 6, configured to calculate the spatial coordinates of the cloud feature points based on the elevation angle and azimuth angle of the cloud feature points and the spatial coordinates of the paired all-sky imagers in a three-dimensional spatial coordinate system;

[0156] The cloud feature coordinate change conversion module 7 is configured to calculate the correlation between the pixel coordinate change and the spatial coordinate change of the same all-sky imager at two adjacent moments according to the spatial coordinates of the cloud feature points;

[0157] The cloud speed calculation module 8 is configured to calculate the cloud movement speed based on the correlation between the pixel coordinate change amount and the space coordinate change amount.

[0158] like Figure 9 As shown, in the above embodiment, preferably, the cloud feature angle calculation module 5 includes:

[0159] The imager angle calculation unit 51 is configured to calculate the azimuth angle β and elevation angle α of each all-sky imager based on the spatial coordinates of the all-sky imager and the coordinate point relationship between the pixel coordinates and the spatial coordinates, specifically:

[0160]

[0161] The cloud feature angle calculation unit 52 is configured to substitute the pixel coordinates of the cloud feature point in the matching feature point set into the azimuth and elevation angles of the all-sky imager to obtain the elevation angle α of the cloud feature point. 1，t，k and α 2，t，k and the azimuth angle β 1，t，k and β 2，t，k .

[0162] like Figure 10 As shown, in the above embodiment, preferably, the cloud feature point space calculation module 6 includes:

[0163] The position relationship calculation unit 61 is configured to obtain the position relationship between the cloud feature point and the paired all-sky imager in the three-dimensional space coordinate system according to the definition of the trigonometric tangent function, which is expressed as follows:

[0164]

[0165] Among them, the spatial coordinates of the paired sky imagers are A1(x1, y1, z1) and A2(x2, y2, z2), and the cloud feature point P c The spatial coordinates (x c ,y c , z c ), the elevation angles α1 and α2 of the paired sky imagers;

[0166] The distance relationship calculation unit 62 is configured to obtain the distance relationship between the cloud feature points and the all-sky imagers in the three-dimensional space coordinate system according to the cosine theorem of the triangle, and the expression is:

[0167] d xy，12 2 =d xy，1c 2 +d xy，2c 2 -2×d xy，1c ×d xy，2c ×cos(|β1-β2|); where d xy，12 Refers to the horizontal distance between two all-sky imagers, d xy，1c and d xy，2c They refer to the horizontal distances between the two all-sky imagers and the cloud feature points;

[0168]

[0169] The cloud feature point spatial calculation unit 63 is configured to calculate the spatial coordinates of the cloud feature points in the three-dimensional space coordinate system based on the elevation angle and azimuth angle, position relationship and distance relationship of the cloud feature points. The expression is:

[0170] in,

[0171]

[0172] Then, according to the coordinate point relationship between pixel coordinates and spatial coordinates, we get:

[0173]

[0174] Finally, the elevation angle α 1，t，k and α 2，t，k , and the azimuth angle β 1，t，k and β 2，t，k Substituting the above formula, we can get the spatial coordinate z of the cloud feature point in the three-dimensional space coordinate system: c，t，k , then z c，t，k Substituting the coordinate point relationship between pixel coordinates and spatial coordinates, we can get the spatial coordinates (x c，t，k ,y c，t，k ).

[0175] In the above embodiment, preferably, the pixel coordinate change is:

[0176] Among them, the pixel coordinates of the spatial feature points captured by any all-sky imager at two adjacent moments are (i1, j1) and (i2, j2); similarly, the kth matching spatial feature point pair P is obtained. 1，t，k (i 1，t，k ,j 1，t，k ) and P 2，t，k (i 2，t，k ,j 2，t，k ) pixel coordinate change Δi t，k and Δj t，k ;

[0177] The change in spatial coordinates is:

[0178]

[0179] i 1，t，k and j 1，t，k , and Δi t，k and Δj t，k and z c，t，k Substituting into the above formula, we can get the spatial coordinate change Δx c，t，k and Δy c，t，k ;

[0180] The expression for cloud movement speed is:

[0181] Where Δt is the time interval between two adjacent moments, so we can get the spatial coordinates of the kth feature point and the cloud moving speed at time t, that is, V c，t，k =(xc，t，k ,y c，t，k , z c，t，k ,u c，t，k , v c，t，k ), all observation data sets at time t are V c，t =(V c，t，1 , V c，t，2 ,…,V c，t，M ), where M is the number of valid cloud feature points.

[0182] Based on the above Figure 5 and Figure 7 The method shown, accordingly, an embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the steps of the cloud position and cloud speed measurement method based on the all-sky cloud map network of any of the above embodiments.

[0183] Based on this understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, USB flash drive, mobile hard disk, etc.), and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods of various implementation scenarios of the present application.

[0184] Based on the above Figures 5 to 7 The method shown, and Figures 8 to 10 In the virtual device embodiment shown, in order to achieve the above-mentioned purpose, the embodiment of the present application also provides a computer device, including a storage medium and a processor; the storage medium is used to store computer programs; the processor is used to execute the computer program to implement the steps of the cloud position and cloud speed measurement method based on the all-sky cloud map network of any of the above-mentioned embodiments.

[0185] Optionally, the computer device may further include a user interface, a network interface, a camera, a radio frequency (RF) circuit, a sensor, an audio circuit, a Wi-Fi module, etc. The user interface may include a display, an input unit such as a keyboard, etc., and the optional user interface may also include a USB interface, a card reader interface, etc. The network interface may optionally include a standard wired interface, a wireless interface (such as a Bluetooth interface, a Wi-Fi interface), etc.

[0186] Those skilled in the art will understand that the computer device structure provided in this embodiment does not constitute a limitation on the computer device, and may include more or fewer components, or a combination of certain components, or different component arrangements.

[0187] The storage medium may also include an operating system and a network communication module. An operating system is a program that manages and stores the hardware and software resources of a computer device, supporting the execution of information processing programs and other software and / or programs. The network communication module facilitates communication between components within the storage medium, as well as with other hardware and software within the physical device.

[0188] In the present invention, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. The term "plurality" refers to two or more, unless expressly limited otherwise. Terms such as "installed," "connected," "connected," and "fixed" should be interpreted broadly. For example, "connected" can mean a fixed connection, a detachable connection, or an integral connection; "connected" can mean a direct connection or an indirect connection through an intermediary. Those skilled in the art will understand the specific meanings of these terms in the present invention based on specific circumstances.

[0189] In the description of the present invention, it should be understood that the directions or positional relationships indicated by terms such as "up", "down", "left", "right", "front" and "back" are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or unit referred to must have a specific direction, be constructed and operated in a specific direction. Therefore, they should not be understood as limiting the present invention.

[0190] Throughout this specification, terms such as "one embodiment," "some embodiments," and "specific embodiments" mean that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0191] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A cloud position and cloud velocity measurement method based on a full sky cloud map network, characterized in that: The following steps are involved: Obtain the full sky image captured by each full sky imager and the relationship between the pixel coordinates of the full sky image and the coordinate points of the spatial coordinates in the three-dimensional spatial coordinate system, specifically: Wherein, the field of view (FOV) of the all-sky imager is 180°, N is the number of pixels in the all-sky image, and the point at the i-th pixel in the horizontal direction and the j-th pixel in the vertical direction of the all-sky image corresponds to the spatial coordinates (x, y, z) of the point P in the three-dimensional spatial coordinate system; Identifying spatial feature matching points of the all-sky image captured by the pairs of all-sky imagers at the same time to obtain spatial feature point pairs; Calculate the direction angle of the line connecting the two points in each pair of spatial feature points, specifically: Among them, the two full sky images I at time t 1，t and I 2，t The coordinates of P are 1，t，k (i 1，t，k ,j 1，t，k ) and P 2，t，k (i 2，t，k ,j 2，t，k ); The spatial feature point pairs are quality-controlled according to a preset threshold corresponding to the direction angle to obtain a matching feature point set. The matching feature point set P F The expression is: P F =[(P 1，t，k , P 2，t，k )ifθ t，k <median(θ t )-10|k=1,2,...,N];where, θ t，k Represented as the kth matching spatial feature point pair P at time t 1，t，k (i 1，t，k ,j 1，t，k ) and P 2，t，k (i 2，t，k ,j 2，t，k )’s connecting line direction angle, median(θ t ) is expressed as the average value of the direction angles of the connecting lines of all spatial feature point pairs at time t. The preset threshold is that the difference between the direction angle of the connecting line of the kth matching spatial feature point pair at time t and the average value of the direction angles of the connecting lines of all spatial feature point pairs at time t is less than 10, that is, θ t，k <median(θ t )-10; Calculating the elevation and azimuth angles of cloud feature points based on the pixel coordinates of the matching feature point set in the full sky image; Calculating the spatial coordinates of the cloud feature point according to the elevation angle and the azimuth angle of the cloud feature point and the spatial coordinates of the paired all-sky imagers in the three-dimensional spatial coordinate system; Calculating the correlation between the pixel coordinate change and the spatial coordinate change of the same all-sky imager at two adjacent moments according to the spatial coordinates of the cloud feature points; The cloud movement speed is calculated based on the correlation between the pixel coordinate change amount and the space coordinate change amount.

2. The cloud position and cloud velocity measurement method according to claim 1, characterized in that: Calculating the elevation angle and the azimuth angle of the cloud feature point according to the pixel coordinates of the matching feature point set in the full sky image comprises the following steps: The azimuth angle β and elevation angle α of each of the all-sky imagers are calculated based on the spatial coordinates of the all-sky imager and the coordinate point relationship between the pixel coordinates and the spatial coordinates, specifically: Substitute the pixel coordinates of the cloud feature point in the matching feature point set into the azimuth and elevation angles of the all-sky imager to obtain the elevation angle α of the cloud feature point 1，t，k and α 2，t，k and the azimuth angle β 1，t，k and β 2，t，k .

3. The cloud position and cloud velocity measurement method according to claim 2, characterized in that: Calculating the spatial coordinates of the cloud feature points according to the elevation angle and the azimuth angle of the cloud feature points and the spatial coordinates of the paired all-sky imagers in the three-dimensional spatial coordinate system comprises the following steps: According to the definition of the trigonometric tangent function, the positional relationship between the cloud feature points and the paired all-sky imagers in the three-dimensional space coordinate system is obtained, and the expression is: Among them, the spatial coordinates of the paired sky imagers are A1(x1, y1, z1) and A2(x2, y2, z2), and the cloud feature point P c The spatial coordinates (x c ,y c , z c ), the elevation angles α1 and α2 of the paired sky imagers; The distance relationship between the cloud feature points and the all-sky imager in the three-dimensional space coordinate system is obtained according to the cosine theorem of the triangle, and the expression is: d xy，12 2 =d xy，1c 2 +d xy，2c 2 -2×d xy，1c ×d xy，2c ×cos(|β1-β2|); where d xy，12 Refers to the horizontal distance between two all-sky imagers, d xy，1c and d xy，2c They refer to the horizontal distances between the two all-sky imagers and the cloud feature points; The spatial coordinates of the cloud feature points in the three-dimensional space coordinate system are calculated based on the elevation angle and azimuth angle of the cloud feature points, the position relationship, and the distance relationship. The expression is: in, Then, according to the coordinate point relationship between pixel coordinates and spatial coordinates, we get: The elevation angle α 1，t，k and α 2，t，k , and the azimuth angle β 1，t，k and β 2，t，k Substituting the above formula, we can get the spatial coordinate z of the cloud feature point in the three-dimensional space coordinate system: c，t，k , then z c，t，k Substituting the pixel coordinates and the spatial coordinates into the coordinate point relationship, the spatial coordinates (x c，t，k ,y c，t，k ).

4. The cloud position and cloud velocity measurement method according to claim 3, characterized in that: The pixel coordinate change is: Among them, the pixel coordinates of the spatial feature points captured by any all-sky imager at two adjacent moments are (i1, j1) and (i2, j2); similarly, the kth matching spatial feature point pair P is obtained. 1，t，k (i 1，t，k ,j 1，t，k ) and P 2，t，k (i 2，t，k ,j 2，t，k ) pixel coordinate change Δi t，k and Δj t，k ; Then, according to the coordinate point relationship between the pixel coordinates and the spatial coordinates, the spatial coordinate change is obtained as: will i 1，t，k and j 1，t，k , and Δi t，k and Δj t，k and z c，t，k Substituting into the above formula, we can get the spatial coordinate change Δx c，t，k and Δy c，t，k ; The cloud movement speed is calculated based on the correlation between the pixel coordinate change and the space coordinate change, specifically: Where Δt is the time interval between two adjacent moments, so we can get the spatial coordinates of the kth feature point and the cloud moving speed at time t, that is, V c，t，k =(x c，t，k ,y c，t，k , z c，t，k ,u c，t，k , v c，t，k ), all observation data sets at time t are V c，t =(V c，t，1 , V c，t，2 ,…,V c，t，M ), where M is the number of valid cloud feature points.

5. A cloud position and cloud velocity measurement system based on a full sky cloud map network, characterized in that: include: The pixel coordinate and spatial coordinate conversion module is configured to obtain the full sky image captured by each full sky imager and the coordinate point relationship between the pixel coordinates of the full sky image and the spatial coordinates in the three-dimensional spatial coordinate system, specifically: Wherein, the field of view (FOV) of the all-sky imager is 180°, N is the number of pixels in the all-sky image, and the point at the i-th pixel in the horizontal direction and the j-th pixel in the vertical direction of the all-sky image corresponds to the spatial coordinates (x, y, z) of the point P in the three-dimensional spatial coordinate system; A spatial feature recognition module is configured to recognize spatial feature matching points of the all-sky image captured by the paired all-sky imagers at the same time, thereby obtaining spatial feature point pairs; The connection direction angle calculation module is configured to calculate the direction angle of the connection line between the two points in each pair of spatial feature points, specifically: Among them, the two full sky images I at time t 1，t and I 2，t The coordinates of P are 1，t，k (i 1，t，k ,j 1，t，k ) and P 2，t，k (i 2，t，k ,j 2，t，k ); The data quality control module is configured to perform quality control on the spatial feature point pair according to a preset threshold value corresponding to the direction angle, and obtain a matching feature point set, the matching feature point set P F The expression is: P F =[(P 1，t，k , P 2，t，k )ifθ t，k <median(θ t )-10|k=1,2,...,N];where, θ t，k Represented as the kth matching spatial feature point pair P at time t 1，t，k (i 1，t，k ,j 1，t，k ) and P 2，t，k (i 2，t，k ,j 2，t，k )’s connecting line direction angle, median(θ t ) is expressed as the average value of the direction angles of the connecting lines of all spatial feature point pairs at time t. The preset threshold is that the difference between the direction angle of the connecting line of the kth matching spatial feature point pair at time t and the average value of the direction angles of the connecting lines of all spatial feature point pairs at time t is less than 10, that is, θ t，k <median(θ t )-10; a cloud feature angle calculation module configured to calculate the elevation angle and azimuth angle of cloud feature points based on the pixel coordinates of the matching feature point set in the full sky image; a cloud feature point spatial calculation module configured to calculate the spatial coordinates of the cloud feature point based on the elevation angle and azimuth angle of the cloud feature point and the spatial coordinates of the paired all-sky imager in the three-dimensional spatial coordinate system; a cloud feature coordinate change conversion module configured to calculate, based on the spatial coordinates of the cloud feature points, a correlation between a pixel coordinate change and a spatial coordinate change of the same all-sky imager at two adjacent moments; The cloud speed calculation module is configured to calculate the cloud movement speed based on the correlation between the pixel coordinate change amount and the space coordinate change amount.

6. The cloud position and cloud velocity measurement system according to claim 5, characterized in that: The cloud feature angle calculation module includes: The imager angle calculation unit is configured to calculate the azimuth angle β and elevation angle α of each of the all-sky imagers according to the spatial coordinates of the all-sky imager and the coordinate point relationship between the pixel coordinates and the spatial coordinates, specifically: A cloud feature angle calculation unit is configured to substitute the pixel coordinates of a cloud feature point in the matching feature point set into the azimuth and elevation angles of the all-sky imager to obtain the elevation angle α of the cloud feature point. 1，t，k and α 2，t，k and the azimuth angle β 1，t，k and β 2，t，k .

7. The cloud position and cloud velocity measurement system according to claim 6, characterized in that: The cloud feature point space calculation module includes: The position relationship calculation unit is configured to obtain the position relationship between the cloud feature point and the paired all-sky imager in the three-dimensional space coordinate system according to the definition of the trigonometric tangent function, and the expression is: Among them, the spatial coordinates of the paired sky imagers are A1(x1, y1, z1) and A2(x2, y2, z2), and the cloud feature point P c The spatial coordinates (x c ,y c , z c ), the elevation angles α1 and α2 of the paired sky imagers; The distance relationship calculation unit is configured to obtain the distance relationship between the cloud feature points and the all-sky imager in the three-dimensional space coordinate system according to the cosine theorem of a triangle, and the expression is: d xy，12 2 =d xy，1c 2 +d xy，2c 2 -2×d xy，1c ×d xy，2c ×cos(|β1-β2|); where d xy，12 Refers to the horizontal distance between two all-sky imagers, d xy，1c and d xy，2c They refer to the horizontal distances between the two all-sky imagers and the cloud feature points; The cloud feature point space calculation unit is configured to calculate the spatial coordinates of the cloud feature point in the three-dimensional space coordinate system based on the elevation angle and azimuth angle of the cloud feature point, the position relationship, and the distance relationship. The expression is: in, Then, according to the coordinate point relationship between pixel coordinates and spatial coordinates, we get: Finally, the elevation angle α 1，t，k and α 2，t，k , and the azimuth angle β 1，t，k and β 2，t，k Substituting the above formula, we can get the spatial coordinate z of the cloud feature point in the three-dimensional space coordinate system: c，t，k , then z c，t，k Substituting the pixel coordinates and the spatial coordinates into the coordinate point relationship, the spatial coordinates (x c，t，k ,y c，t，k ).

8. The cloud position and cloud velocity measurement system according to claim 7, characterized in that: The pixel coordinate change is: Among them, the pixel coordinates of the spatial feature points captured by any all-sky imager at two adjacent moments are (i1, j1) and (i2, j2); similarly, the kth matching spatial feature point pair P is obtained. 1，t，k (i 1，t，k ,j 1，t，k ) and P 2，t，k (i 2，t，k ,j 2，t，k ) pixel coordinate change Δi t，k and Δj t，k ; The spatial coordinate variation is: i 1，t，k and j 1，t，k , and Δi t，k and Δj t，k and z c，t，k Substituting into the above formula, we can get the spatial coordinate change Δx c，t，k and Δy c，t，k ; The expression for cloud movement speed is: Where Δt is the time interval between two adjacent moments, so we can get the spatial coordinates of the kth feature point and the cloud moving speed at time t, that is, V c，t，k =(x c，t，k ,y c，t，k , z c，t，k ,u c，t，k , v c，t，k ), all observation data sets at time t are V c，t =(V c，t，1 , V c，t，2 ,…,V c，t，M ), where M is the number of valid cloud feature points.

9. A readable storage medium, characterized in that: A computer program is stored thereon, which, when executed by a processor, implements the steps of the cloud position and cloud speed measurement method based on the all-sky cloud map network as described in any one of claims 1 to 4.

10. A computer device, characterized in that: The invention comprises a storage medium and a processor; the storage medium is used to store a computer program; the processor is used to execute the computer program to implement the steps of the cloud position and cloud speed measurement method based on the all-sky cloud map network as described in any one of claims 1 to 4.

Images