Base station antenna working parameter measurement method, device, medium and computer program product

Abstract

The invention relates to the technical field of communication, and particularly provides a base station antenna working parameter measurement method and device, a medium and a computer program product. The base station antenna working parameter measurement method comprises the following steps: shooting a picture of a to-be-detected antenna through an unmanned aerial vehicle, and obtaining working parameters of the unmanned aerial vehicle when the picture is shot; inputting the picture into a preset antenna attitude detection model to obtain an attitude type of the to-be-detected antenna and a rotation angle of the rectangular detection frame; the rectangular detection frame is used for positioning the outer contour of a to-be-detected antenna; and determining the working parameter data of the to-be-detected antenna based on the working parameters of the unmanned aerial vehicle, the attitude type and the rotation angle of the rectangular detection frame. The antenna attitude type and the working parameter data are directly acquired by using the unmanned aerial vehicle, so that the requirement of manual measurement is reduced, the efficiency and the precision of data acquisition are improved, the personal error in data processing is reduced, and the universality and the adaptability of the technology are enhanced.

Description

Technical Field

[0001] The present disclosure relates to the field of communication technology, and in particular to a method, device, medium and computer program product for measuring engineering parameters of a base station antenna. Background Art

[0002] With the continuous development of wireless communication technology, the amount of business undertaken by base station antennas has increased significantly. Therefore, accurate detection and optimization of the engineering parameter data of base station antennas has become an important direction to promote the development of wireless communication services. Among them, the engineering parameter data of base station antennas include the latitude and longitude, altitude, azimuth angle and downtilt angle of the antenna.

[0003] In related technologies, the latitude and longitude information of the base station antenna is often obtained by using a GPS (Global Positioning System) measuring instrument, and the height of the base station antenna is measured using a rangefinder, while the measurement of the azimuth angle and downtilt angle can only be obtained based on design drawings or acceptance documents. This method that relies on non-real-time data results in low measurement accuracy of the engineering parameter data, thus affecting the performance and service quality of the entire wireless network. Summary of the invention

[0004] In view of this, exemplary embodiments of the present disclosure provide a method, device, medium and computer program product for measuring engineering parameters of a base station antenna to solve the problems existing in the related art.

[0005] In one aspect of an exemplary embodiment of the present disclosure, a method for measuring engineering parameters of a base station antenna is provided, the method comprising:

[0006] Taking pictures of the antenna to be tested by a drone, and obtaining the working parameters of the drone when taking the pictures;

[0007] Input the image into a preset antenna posture detection model to obtain the posture type of the antenna to be detected and the rotation angle of a rectangular detection frame; the rectangular detection frame is used to locate the outer contour of the antenna to be detected;

[0008] Based on the working parameters of the UAV, the posture type and the rotation angle of the rectangular detection frame, the working parameter data of the antenna to be detected is determined.

[0009] Another aspect of the exemplary embodiments of the present disclosure provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the method described in the exemplary embodiments of the present disclosure.

[0010] According to another aspect of the exemplary embodiments of the present disclosure, a computer-readable storage medium is provided, on which a computer program / instruction is stored. When the computer program / instruction is executed by a processor, the method described in the exemplary embodiments of the present disclosure is implemented.

[0011] According to another aspect of the exemplary embodiments of the present disclosure, a computer program product is provided, including a computer program / instruction. When the computer program / instruction is executed by a processor, the method described in the exemplary embodiments of the present disclosure is implemented.

[0012] As will be described in detail below, according to a method for measuring the engineering parameters of a base station antenna according to an embodiment of the present disclosure, a picture of the antenna to be detected is taken by a drone, and the working parameters of the drone when taking the picture are obtained; the picture is input into a preset antenna attitude detection model to obtain the attitude type of the antenna to be detected and the rotation angle of the rectangular detection frame; the rectangular detection frame is used to locate the outer contour of the antenna to be detected; based on the working parameters of the drone, the attitude type and the rotation angle of the rectangular detection frame, the engineering parameter data of the antenna to be detected is determined. Therefore, the method for measuring the engineering parameters of the base station antenna provided by the present disclosure takes pictures of the antenna to be detected from different angles by a drone, and records the working parameters of the drone during the shooting process. The pictures taken are then input into a pre-trained antenna attitude detection model. The antenna attitude detection model can identify the attitude type of the antenna and calculate the rotation angle of the rectangular detection frame that locates the outer contour of the antenna. The rotation angle represents the angle of the rectangular detection frame relative to the vertical direction. Furthermore, according to the working parameters of the drone, the attitude type and the rotation angle of the rectangular detection frame, the engineering parameter data of the antenna to be detected can be accurately determined, thereby improving the measurement accuracy of the antenna engineering parameter data. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The above and other purposes, features and advantages of the present disclosure will become more apparent by describing the embodiments of the present disclosure in more detail in conjunction with the accompanying drawings. The accompanying drawings are used to provide a further understanding of the embodiments of the present disclosure and constitute a part of the specification. Together with the embodiments of the present disclosure, they are used to explain the present disclosure and do not constitute a limitation of the present disclosure. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0014] Figure 1 A flowchart of a method for measuring engineering parameters of a base station antenna provided by an exemplary embodiment of the present disclosure;

[0015] Figure 2 This is a schematic diagram of antenna image annotation provided as an example in the present disclosure;

[0016] Figure 3 This is a schematic diagram of a horizontal rectangular frame IOU provided as an example in the present disclosure;

[0017] Figure 4This is a schematic diagram of an IOU of a rotating rectangular frame provided as an example in the present disclosure;

[0018] Figure 5 A flow chart of a method for measuring engineering parameters of a base station antenna provided by an exemplary embodiment of the present disclosure;

[0019] Figure 6 A schematic block diagram of functional modules of a base station antenna engineering parameter measurement device provided as an example in the present disclosure;

[0020] Figure 7 A structural block diagram of an electronic device provided as an example in the present disclosure;

[0021] Figure 8 A schematic diagram of a computer program product provided by an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0022] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0023] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0024] The term "including" and its variations used in this document are open inclusions, that is, "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments". Relevant definitions of other terms will be given in the description below. It should be noted that the concepts of "first", "second", etc. mentioned in this disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0025] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0026] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0027] It is understandable that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, scope of use, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0028] For example, in response to receiving an active request from a user, a prompt message is sent to the user to clearly prompt the user that the operation requested to be performed will require obtaining and using the user's personal information. Thus, the user can autonomously choose whether to provide personal information to software or hardware such as an electronic device, application, server, or storage medium that performs the operation of the technical solution of the present disclosure according to the prompt message.

[0029] As an optional but non-limiting implementation, in response to receiving an active request from the user, the method of sending a prompt message to the user may be, for example, a pop-up window, in which the prompt message may be presented in text. In addition, the pop-up window may also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device. It is understandable that the above notification and the process of obtaining user authorization are only illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that meet relevant laws and regulations may also be applied to the implementation of the present disclosure.

[0030] With the continuous development of wireless communication technology, the amount of business undertaken by base station antennas has increased significantly. Therefore, accurate detection and optimization of the engineering parameter data of base station antennas has become an important direction to promote the development of wireless communication services. Among them, the engineering parameter data of base station antennas include the latitude and longitude, altitude, azimuth angle and downtilt angle of the antenna.

[0031] In related technologies, the latitude and longitude information of the base station antenna is often obtained by using a GPS measuring instrument, and the height of the base station antenna is measured using a rangefinder, while the measurement of the azimuth angle and downtilt angle can only be obtained based on design drawings or acceptance documents. This method that relies on non-real-time data results in low measurement accuracy of the engineering parameter data, thus affecting the performance and service quality of the entire wireless network.

[0032] Therefore, in order to solve the above problems, an exemplary embodiment of the present disclosure provides a method for measuring the engineering parameters of base station antennas, in which the front and side faces of each antenna are photographed by a drone, and the antenna attitude is detected for the photographed pictures. If the target antenna in the photo is detected as facing the front, the station height and longitude and latitude of the antenna can be obtained by reading the altitude and longitude and latitude of the drone, and the azimuth of the camera at this time can be read. After flipping 180°, the direction angle of the antenna can be obtained. If the target antenna in the photo is detected as facing the side, the downtilt angle of the antenna is obtained by the rotation angle result of the rectangular detection frame. If the target antenna in the photo is detected at other angles, that is, not facing the front or the side, it will not be processed. After obtaining the station height, longitude and latitude, direction angle and downtilt angle of the target antenna, the measurement of the antenna engineering parameters is completed.

[0033] For example, Figure 1 This is a flow chart of a method for measuring engineering parameters of a base station antenna provided by an exemplary embodiment of the present disclosure, such as Figure 1 As shown, the following steps may be specifically included:

[0034] Step S110: Acquire a picture of the antenna to be detected.

[0035] First, control the drone to fly near the antenna and take side shots. The specific shooting requirements include:

[0036] Make sure the drone is flying at approximately the same height as the antenna, and adjust the roll and pitch angles of the camera to stay close to 0 degrees so that it can be aimed horizontally to the side of the antenna.

[0037] Keep the distance between the drone and the antenna between 10 meters and 30 meters. The focal length of the camera should be adjusted to focus on the center of the side of the antenna to avoid image distortion caused by shooting.

[0038] Make sure only the side of the antenna is included in the frame and avoid photographing the other 5 sides of the antenna.

[0039] After shooting the side, move the drone to the front of the antenna for shooting. Specific shooting requirements include:

[0040] Position the drone in front of the antenna and keep an appropriate shooting distance of 10 to 30 meters from the antenna.

[0041] The focal length of the camera should be adjusted to focus on the center of the front of the antenna to avoid image distortion caused by shooting.

[0042] Make sure that only the front side of the antenna is included in the picture, and avoid photographing the other 5 sides of the antenna.

[0043] Based on this, according to the above shooting requirements, the drone can effectively capture detailed images of the antenna from different angles, which can greatly improve the accuracy and efficiency of the measurement. These images are subsequently used to analyze the specific parameters of the antenna, such as azimuth angle, downtilt angle, etc.

[0044] Step S120: The image of the antenna to be detected is input into the antenna posture detection model to obtain the posture type of the antenna to be detected.

[0045] The antenna attitude detection model is used to detect the antenna attitude of the image to be detected and determine the attitude type. The attitude types include: front, side, and other attitudes.

[0046] Exemplarily, the antenna attitude detection model can be built based on the YOLOv5 (You Only Look Once version 5) framework. YOLOv5 is an efficient and high-precision single-stage target detection algorithm. The algorithm regards the target detection task as a classification plus regression problem, and directly predicts the bounding box and category of the target through a convolutional neural network. The advantage of the YOLOv5 framework is that it has a fast detection speed while ensuring high accuracy, which is suitable for real-time application scenarios.

[0047] Exemplarily, the training process of the antenna attitude detection model may include:

[0048] First, prepare the training data.

[0049] The training data can include antenna posture image data, including more than 1,000 front and side shots. Using drones to collect data according to previously set shooting strategies can ensure data diversity and coverage, including different types of antennas, different weather conditions, and diverse background environments. The preparation of training data is to provide sufficient and representative training samples for subsequent model training.

[0050] Label the training data. The labeling process involves using a rotatable detection box to locate the antenna in the image and classify it: antennas facing the front are labeled "1", antennas facing the side are labeled "2", and antennas shot from other angles are labeled "0". The labeling not only involves the category of the antenna, but also includes its position coordinates in the image, the length and width of the detection box, and because the detection box is rotatable, the rotation angle of the detection box must also be recorded. This detailed labeling work is the basis for the model to learn how to accurately identify and classify antenna postures.

[0051] For example, Figure 2 This is an exemplary antenna image annotation schematic diagram provided by the present disclosure, such as Figure 2As shown in the figure, the black solid line frame represents the antenna outline, the black dotted line frame represents the detection frame, the antenna facing the front is marked as "1", the antenna facing the side is marked as "2", and the antenna shot at other angles is marked as "0".

[0052] Then, the antenna attitude detection model is trained.

[0053] In order to make the YOLOv5 model adapt to the special needs of antenna attitude detection, in this embodiment, a series of key adjustments are made to its detection frame and related calculations. These adjustments mainly involve the format of the model output and the definition of the loss function, including:

[0054] Adjustments to model output:

[0055] The original YOLOv5 model output includes the location, size, classification category probability (such as front, side, or other), and confidence score of each detection box.

[0056] In this embodiment, the model output of YOLOv5 after the improved detection frame is the position, size, classification category probability (such as front, side or other), confidence score and rotation angle of the detection frame relative to the center of each detection frame. The rotation angle of the detection frame relative to the center is the angle between the center line of the detection frame and the vertical direction, which can characterize the rotation state of the detection frame and is the key data for accurately judging the posture of the antenna.

[0057] Adjustment of loss function:

[0058] The original YOLOv5 loss function mainly includes bounding box loss, classification loss and confidence loss. In the application scenario of antenna attitude detection, the classification loss and confidence loss remain unchanged, but the bounding box loss needs to be adjusted to adapt to the rotation detection frame.

[0059] The bounding box loss function is the CIoU (Complete Intersection over Union) loss function. The traditional CIoU loss calculation is based on the intersection over union (IoU) of the bounding box, which is an important indicator to measure the closeness between the predicted bounding box and the true bounding box. The IOU is defined as the ratio of the intersection area to the union area of ​​two bounding boxes. The larger the IoU value, the closer the predicted detection box is to the true label, that is, the more accurate the detection result.

[0060] For example, Figure 3 This is a schematic diagram of a horizontal rectangular frame IOU provided as an example in the present disclosure. Figure 3 As shown in the figure, the area of ​​the overlapping part of the detection box of the label (ground truth) and the predicted detection box is S12, and the areas of the non-overlapping parts are S1 and S2 respectively. Then, the calculation formula of IoU can be expressed as:

[0061]

[0062] Among them, S12 represents the overlapping area of ​​the label detection frame and the predicted detection frame; S1 represents the non-overlapping area of ​​the label detection frame; S2 represents the non-overlapping area of ​​the predicted detection frame.

[0063] However, the standard IoU only works for axis-aligned rectangular boxes. In order to handle rotated detection boxes, new geometric calculations need to be introduced to accurately calculate the IoU between rotated boxes.

[0064] For example, Figure 4 This is a schematic diagram of an IOU of a rotating rectangular frame provided by an exemplary embodiment of the present disclosure, such as Figure 4 As shown in the figure, the area of ​​the overlapping part of the detection box of the label (ground truth) and the predicted detection box is S12, and the areas of the non-overlapping parts are S1 and S2 respectively. When the detection box is rotated to a certain extent, the IoU calculation not only considers the position and size of the box, but also the rotation angle of the box. In this embodiment, the opencv function can be used to solve the IoU of the rotated rectangular box.

[0065] Exemplarily, it is assumed that the labeled (ground truth) rectangular detection box is Box1 = [x0, y0, w0, h0, theta0], and the predicted (predicted) rectangular detection box is Box2 = [x, y, w, h, theta], wherein x0 and y0 represent the horizontal and vertical coordinates of the upper left corner vertex of the labeled rectangular box, respectively; w0 and h0 represent the width and height of the labeled rectangular box, respectively; theta0 represents the rotation angle of the labeled rectangular box relative to the center; x and y represent the horizontal and vertical coordinates of the upper left corner vertex of the predicted rectangular box, respectively; w and h represent the width and height of the predicted rectangular box, respectively; theta represents the rotation angle of the predicted rectangular box relative to the center.

[0066] Furthermore, the areas of the two rectangular boxes are calculated according to the width and height of the rectangular boxes.

[0067] The area of ​​the label rectangular detection box is area1 = w0×h0.

[0068] The area of ​​the predicted rectangular detection box is area2 = w × h.

[0069] Furthermore, the intersection of the two rotated rectangles is calculated.

[0070] For example, the cv2.rotatedRectangleIntersection function can be used to calculate the intersection of two rotated rectangles. If the two rotated rectangles have an intersection, a point set representing the intersection area is returned, and the point set includes the position information of all vertices in the intersection area. If the two rotated rectangles do not have an intersection, the point set is empty.

[0071] Then, the area of ​​the intersection region is calculated.

[0072] If the two rotated rectangles have no intersection, the IoU is 0.

[0073] If there is an intersection between the two rotated rectangles, you can use the cv2.convexHull function to sort the vertices to ensure that the area enclosed by the vertices is a convex polygon. Then, use cv2.contourArea to calculate the area of ​​the convex hull, that is, the area of ​​the intersection area S12.

[0074] The area of ​​the union region is the sum of the areas of the two rectangles minus the area of ​​the intersection region, that is, area1+area2-S12.

[0075] Then, the IoU calculation formula of the rotated rectangle can be expressed as:

[0076]

[0077] Among them, S12 represents the overlapping area of ​​the label detection box and the predicted detection box; S1 represents the non-overlapping area of ​​the label detection box; S2 represents the non-overlapping area of ​​the predicted detection box; area1 represents the area of ​​the label rectangular detection box; area2 represents the area of ​​the predicted rectangular detection box.

[0078] Based on this, this IoU calculation takes into account not only the position and size of the box, but also the rotation angle of the box, which can effectively handle the overlapping area of ​​​​the two rotated rectangles and ensure that the loss function can accurately reflect the geometric difference between the predicted box and the true box, thereby guiding the model to adjust its predictions more accurately during training.

[0079] In addition, through the above adjustments, the improved YOLOv5 model will be able to better adapt to the antenna attitude detection task and provide more accurate detection results, thus meeting the dual requirements of high precision and real-time in practical applications.

[0080] Finally, the trained antenna attitude detection model can be deployed. The deployment steps may include:

[0081] First, build a Docker image. Package the trained YOLOv5 model, necessary test code, and all dependent library functions into a Docker image.

[0082] Then, deploy the image to the server. Deploy the built Docker image to the server and activate the model service by running the Docker container. The environment in the Docker container is exactly the same as the image, which can ensure the stability and efficiency of the model operation.

[0083] Then, external requests are processed through the web server on the server. A web API can be designed to receive image data uploaded by users. Users only need to send images to the specified API endpoint through HTTP requests, and the web server will parse the data and pass the images to the detection model running in the Docker container.

[0084] After receiving the image, the YOLOv5 model in the Docker container starts reasoning, identifying the objects in the image and analyzing their posture. After reasoning is completed, the detection results are immediately sent back to the web server and then returned to the user in real time through the API in JSON format or other predetermined format.

[0085] Step S130: When the antenna posture is frontal, determine the azimuth angle according to the camera parameters.

[0086] If the antenna attitude is detected to be forward, read the current azimuth of the drone camera, and the opposite direction of the azimuth is the direction angle of the target antenna.

[0087] Step S140: When the antenna posture is sideways, determine the downtilt angle according to the detection frame.

[0088] If the antenna posture is detected to be sideways, the rotation angle of the center of the detection frame (ie, theta) is obtained, and the rotation angle of the center of the detection frame is the downtilt angle of the target antenna.

[0089] Step S150: When the antenna posture is other postures, no processing is performed.

[0090] If it is detected that the antenna posture is in other postures, that is, the antenna is not facing the front or the side, the antenna image will not be processed.

[0091] Complete the engineering parameter measurement of all base station antennas in sequence and obtain the engineering parameter data of all antennas.

[0092] One or more technical solutions provided in the exemplary embodiments of the present disclosure control the drone to fly near the antenna, and photograph the front and side of each antenna one by one; perform antenna posture detection on the photographed photos. If the target antenna in the photo is detected as facing the front, the station height and longitude and latitude of the antenna can be obtained by reading the altitude and longitude and latitude of the drone. At the same time, read the azimuth of the camera at this time, and the direction angle of the antenna can be obtained after flipping 180°. If the target antenna in the photo is detected as facing the side, the downtilt angle of the target antenna is obtained by the rotation angle result of the rectangular detection frame. If the target antenna in the photo is detected at other angles, that is, not facing the front or the side, it will not be processed. After completing the station height, longitude and latitude, direction angle and downtilt angle of the target antenna, the measurement of the antenna engineering parameters is completed.

[0093] Among them, YOLOv5 is used to detect the antenna posture when facing forward and sideways. For the side-facing one, the rotation angle of the detection frame is also output. YOLOv5 optimizes the bounding box loss function and determines the rotation angle by calculating the IoU of the rotated rectangular box.

[0094] Therefore, the base station antenna engineering parameter measurement method provided in the exemplary embodiment of the present disclosure accurately matches the outer contour of the antenna by rotating the detection frame, thereby providing accurate antenna engineering parameter data.

[0095] Based on the above embodiments, the present disclosure also provides a method for measuring engineering parameters of a base station antenna. Figure 5 This is a flow chart of a method for measuring the working parameters of a base station antenna provided by an exemplary embodiment of the present disclosure, such as Figure 5 As shown, the method may include the following steps:

[0096] Step S510: Use a drone to take a picture of the antenna to be tested, and obtain the working parameters of the drone when taking the picture.

[0097] In the embodiment, the drone is first controlled to fly near the antenna to take side shots. The specific shooting requirements include:

[0098] Make sure the drone is flying at approximately the same height as the antenna, and adjust the roll and pitch angles of the camera to stay close to 0 degrees so that it can be aimed horizontally to the side of the antenna.

[0099] Keep the distance between the drone and the antenna between 10 meters and 30 meters. The focal length of the camera should be adjusted to focus on the center of the side of the antenna to avoid image distortion caused by shooting.

[0100] Make sure only the side of the antenna is included in the frame and avoid photographing the other 5 sides of the antenna.

[0101] After shooting the side, move the drone to the front of the antenna for shooting. Specific shooting requirements include:

[0102] Position the drone in front of the antenna and keep an appropriate shooting distance of 10 to 30 meters from the antenna.

[0103] The focal length of the camera should be adjusted to focus on the center of the front of the antenna to avoid image distortion caused by shooting.

[0104] Make sure that only the front side of the antenna is included in the picture, and avoid photographing the other 5 sides of the antenna.

[0105] Exemplarily, the drone operating parameters may include: the altitude, longitude and latitude during shooting, and the azimuth angle of the camera, etc.

[0106] Based on this, according to the above shooting requirements, the drone can effectively capture detailed images of the antenna from different angles, which can greatly improve the accuracy and efficiency of the measurement. These images are subsequently used to analyze the specific parameters of the antenna, such as azimuth angle, downtilt angle, etc.

[0107] Step S520: input the image into a preset antenna posture detection model to obtain the posture type of the antenna to be detected and the rotation angle of the rectangular detection frame; the rectangular detection frame is used to locate the outer contour of the antenna to be detected.

[0108] In the embodiment, the antenna posture detection model is used to perform antenna posture detection on the antenna image to be detected and determine the posture type. The posture types include: front, side and other postures.

[0109] Exemplarily, the antenna attitude detection model can be built based on the YOLOv5 (You Only Look Once version 5) framework. YOLOv5 is an efficient and high-precision single-stage target detection algorithm. The algorithm regards the target detection task as a classification plus regression problem, and directly predicts the bounding box and category of the target through a convolutional neural network. The advantage of the YOLOv5 framework is that it has a fast detection speed while ensuring high accuracy, which is suitable for real-time application scenarios.

[0110] Exemplarily, the training process of the antenna attitude detection model may include:

[0111] First, prepare the training data.

[0112] The training data can include antenna posture image data, including more than 1,000 front and side shots. Using drones to collect data according to previously set shooting strategies can ensure data diversity and coverage, including different types of antennas, different weather conditions, and diverse background environments. The preparation of training data is to provide sufficient and representative training samples for subsequent model training.

[0113] Label the training data. The labeling process involves using a rotatable detection box to locate the antenna in the image and classify it: antennas facing the front are labeled "1", antennas facing the side are labeled "2", and antennas shot from other angles are labeled "0". The labeling not only involves the category of the antenna, but also includes its position coordinates in the image, the length and width of the detection box, and because the detection box is rotatable, the rotation angle of the detection box must also be recorded. This detailed labeling work is the basis for the model to learn how to accurately identify and classify antenna postures.

[0114] Then, the antenna attitude detection model is trained.

[0115] In order to make the YOLOv5 model adapt to the special needs of antenna attitude detection, in this embodiment, a series of key adjustments are made to its detection frame and related calculations, including:

[0116] Adjustments to model output:

[0117] The original YOLOv5 model output includes the location, size, classification category probability (such as front, side, or other), and confidence score of each detection box.

[0118] In this embodiment, the model output of YOLOv5 after the improved detection frame is the position, size, classification category probability (such as front, side or other), confidence score and rotation angle of the detection frame relative to the center of each detection frame. The rotation angle of the detection frame relative to the center is the angle between the center line of the detection frame and the vertical direction, which can characterize the rotation state of the detection frame and is the key data for accurately judging the posture of the antenna.

[0119] Step S530: Determine the working parameter data of the antenna to be detected based on the working parameters of the drone, the posture type, and the rotation angle of the rectangular detection frame.

[0120] In the embodiment, when the drone photographs the antenna, it will record a variety of operating parameters, including but not limited to the drone's position (latitude, longitude and altitude), shooting angle, timestamp, etc. The drone's operating parameters are crucial for subsequently determining the precise position and direction of the antenna. For example, the drone's position and altitude can help calculate the absolute height of the antenna, while the shooting angle is related to the antenna's azimuth.

[0121] The rotation angle of the rotating rectangular detection frame reflects the angle of the antenna relative to the vertical direction. By analyzing the rotation angle of the rectangular detection frame, the downtilt angle of the antenna can be inferred.

[0122] The attitude types include: front, side, and other attitudes. The attitude type can reflect the orientation of the antenna. By combining the information of the altitude of the drone and the downtilt angle of the antenna, the specific position of the antenna facing the ground can be accurately calculated. In addition, the timestamp information helps to record and track the status changes of the antenna, which is particularly important for time series analysis.

[0123] Based on this, the drone takes pictures of the antenna to be tested from different angles, and the drone working parameters are recorded during the shooting process. The pictures taken are then input into a pre-trained antenna attitude detection model. The antenna attitude detection model can identify the attitude type of the antenna and calculate the rotation angle of the rectangular detection frame that locates the outer contour of the antenna. The rotation angle represents the angle of the rectangular detection frame relative to the vertical direction. Furthermore, according to the drone working parameters, attitude type and rotation angle of the rectangular detection frame, the engineering parameter data of the antenna to be tested can be accurately determined, thereby improving the measurement accuracy of the antenna engineering parameter data.

[0124] Based on the above embodiment, in another embodiment provided by the present disclosure, the loss function of the above antenna attitude detection model includes a bounding box loss function, which is used to measure the proximity between the predicted rectangular detection box and the real bounding box. The loss value of the bounding box loss function is determined based on the ratio of the intersection area and the union area between the rectangular detection box and the real bounding box, and may include:

[0125] In an embodiment, the adjustment of the model also includes: adjustment of the loss function.

[0126] The original YOLOv5 loss function mainly includes bounding box loss, classification loss and confidence loss. In the application scenario of antenna attitude detection, the classification loss and confidence loss remain unchanged, but the bounding box loss needs to be adjusted to adapt to the rotation detection frame.

[0127] The bounding box loss function is the CIoU (Complete Intersection over Union) loss function. The traditional CIoU loss calculation is based on the intersection over union (IoU) of the bounding box, which is an important indicator to measure the closeness between the predicted bounding box and the true bounding box. The IOU is defined as the ratio of the intersection area to the union area of ​​two bounding boxes. The larger the IoU value, the closer the predicted detection box is to the true label, that is, the more accurate the detection result.

[0128] Based on this, by adjusting the bounding box loss according to the needs of the rotating detection box, the model's ability to recognize antennas at different angles is enhanced, so that antenna parameters can be measured and analyzed more accurately. This optimization makes the model more suitable for antenna detection tasks in complex environments and improves the reliability and efficiency of practical applications.

[0129] Based on the above embodiment, in another embodiment provided by the present disclosure, the loss value of the above bounding box loss function is determined based on the ratio of the intersection area and the union area between the rectangular detection box and the true bounding box, which may include:

[0130] Get the location information of the real bounding box and the rectangular detection box respectively;

[0131] Based on the position information, the area of ​​the real bounding box, the area of ​​the rectangular detection box and the area of ​​the intersection area are determined respectively;

[0132] Determine the area of ​​the union region based on the area of ​​the true bounding box, the area of ​​the rectangular detection box, and the area of ​​the intersection region;

[0133] Based on the area of ​​the intersection region and the area of ​​the union region, the loss value of the bounding box loss function is determined.

[0134] In the embodiment, the loss value of the bounding box loss function is calculated by determining the position information of two rectangular boxes (the predicted rectangular detection box and the real bounding box) to calculate their respective areas and their intersection and union areas. The intersection area reflects the overlapping part of the predicted box and the real box, while the union area includes the entire coverage area of ​​the two boxes.

[0135] The calculation of the loss value is based on the intersection over union (IoU), which is a measure of the similarity between two bounding boxes. The IoU value is determined by the ratio of the area of ​​the intersection area to the area of ​​the union area. The higher the IoU value, the higher the overlap between the predicted bounding box and the true box, that is, the higher the prediction accuracy of the model.

[0136] Based on this, using IoU as the loss function can intuitively evaluate the performance of the detection model. It can not only accurately measure the position error of the two rectangular boxes (the predicted rectangular detection box and the real bounding box), but also consider the accuracy of the size, shape and orientation of the detection box, making the model training closer to the accuracy requirements of actual applications.

[0137] Based on the above embodiment, in another embodiment provided by the present disclosure, respectively determining the real bounding box area, the rectangular detection box area and the intersection area based on the position information may include:

[0138] Get the vertex position information, width, height and rotation angle of the real bounding box;

[0139] Determine the area of ​​the true bounding box based on the width and height of the true bounding box;

[0140] Get the vertex position information, width, height and rotation angle of the rectangular detection box;

[0141] Determine the area of ​​the rectangular detection frame based on the width and height of the rectangular detection frame;

[0142] The area of ​​the intersection region is determined based on the vertex position information and rotation angle of the real bounding box and the vertex position information and rotation angle of the rectangular detection box.

[0143] In the embodiment, it is assumed that the real bounding box is Box1 = [x0, y0, w0, h0, theta0], and the rectangular detection box is Box2 = [x, y, w, h, theta], wherein x0 and y0 respectively represent the horizontal and vertical coordinates of the upper left corner vertex of the real bounding box; w0 and h0 respectively represent the width and height of the real bounding box; theta0 represents the rotation angle of the real bounding box relative to the center; x and y respectively represent the horizontal and vertical coordinates of the upper left corner vertex of the rectangular detection box; w and h respectively represent the width and height of the rectangular detection box; theta represents the rotation angle of the rectangular detection box relative to the center.

[0144] Furthermore, the areas of the two rectangular boxes are calculated according to the width and height of the rectangular boxes.

[0145] The area of ​​the ground-truth bounding box is area1 = w0 × h0.

[0146] The area of ​​the rectangular detection frame is area2=w×h.

[0147] Furthermore, the intersection of the two rotated rectangles is calculated.

[0148] For example, the cv2.rotatedRectangleIntersection function can be used to calculate the intersection of two rotated rectangles. If the two rotated rectangles have an intersection, a point set representing the intersection area is returned, and the point set includes the position information of all vertices in the intersection area. If the two rotated rectangles do not have an intersection, the point set is empty.

[0149] Then, the area of ​​the intersection region is calculated.

[0150] If the two rotated rectangles have no intersection, the IoU is 0.

[0151] If there is an intersection region between the two rotated rectangles, the area of ​​the intersection region may be determined based on the vertex position information and the rotation angle of the real bounding box and the vertex position information and the rotation angle of the rectangular detection box.

[0152] Based on this, this IoU calculation takes into account not only the position and size of the box, but also the rotation angle of the box, which can effectively handle the overlapping area of ​​​​the two rotated rectangles and ensure that the loss function can accurately reflect the geometric difference between the predicted box and the true box, thereby guiding the model to adjust its predictions more accurately during training.

[0153] Based on the above embodiment, in another embodiment provided by the present disclosure, the above determination of the intersection area based on the vertex position information and rotation angle of the real bounding box and the vertex position information and rotation angle of the rectangular detection box may include:

[0154] Based on the vertex position information and rotation angle of the real bounding box and the vertex position information and rotation angle of the rectangular detection box, the intersection of the real bounding box and the rectangular detection box is detected to obtain multiple vertex position information of the intersection area;

[0155] Sorting multiple vertex position information to form a closed polygon;

[0156] Divide the polygon into a number of triangles, and calculate the areas corresponding to the triangles based on multiple vertex position information;

[0157] Accumulate the areas corresponding to several triangles respectively, and use the accumulated results as the area of ​​the intersection area.

[0158] In the embodiment, first, based on the vertex position information and rotation angle of each bounding box, the intersection of the two bounding boxes is detected to determine the multiple vertices of the intersection area. This step is a prerequisite for calculating the intersection area, because only when the specific shape of the intersection area is determined can the area be further calculated.

[0159] Then, sort these vertices to ensure that the area enclosed by the vertices is a closed convex polygon.

[0160] Furthermore, in order to accurately calculate the area of ​​this polygon, it is divided into several triangles. The area of ​​each triangle can be calculated by geometric methods, such as using the vector cross product formula. The areas of these triangles are added up to finally get the total area of ​​the entire intersection polygon.

[0161] In an optional way, you can use cv2.rotatedRectangleIntersection function to calculate the intersection of two bounding boxes. If the two bounding boxes have an intersection, a point set representing the intersection area is returned, which includes the position information of all vertices in the intersection area. If the two bounding boxes do not have an intersection, the point set is empty.

[0162] Then, the area of ​​the intersection region is calculated.

[0163] If two bounding boxes have no intersection, the IoU is 0.

[0164] If there is an intersection area between the two bounding boxes, you can use the cv2.convexHull function to sort the vertices to ensure that the area enclosed by the vertices is a convex polygon. Then, use cv2.contourArea to calculate the area of ​​the convex hull, that is, the area of ​​the intersection area.

[0165] Based on this, the intersection area can be accurately determined by the above method, and then the loss value of the bounding box loss function can be determined. The antenna attitude detection model is updated in reverse according to the loss value, so that the rectangular detection box predicted by the model is more accurate.

[0166] Based on the above embodiment, in another embodiment provided by the present disclosure, the above drone operating parameters include the drone altitude, longitude and latitude, and camera azimuth. The above step S530 may include:

[0167] The altitude and longitude and latitude of the drone are determined as the station altitude and longitude and latitude of the antenna to be tested;

[0168] When the posture type is front-facing, the camera azimuth is obtained, and the camera azimuth is flipped 180 degrees to determine it as the azimuth of the antenna to be detected.

[0169] In the embodiment, when using a drone for antenna detection and attitude analysis, the flight altitude and longitude and latitude data of the drone can be directly used to determine the station altitude and longitude and latitude of the antenna to be detected. This method directly extracts the exact location information of the antenna from the data obtained by the drone, without the need for additional ground measurements, greatly saving time and resources.

[0170] In addition, considering the importance of the camera's azimuth in antenna attitude analysis, the azimuth of the antenna to be detected can be determined by obtaining the azimuth of the drone camera and flipping it 180 degrees. This is because, based on the situation that the drone camera is facing the front of the antenna, that is, the drone camera points to the back of the antenna, flipping the camera's azimuth by 180 degrees is the azimuth of the antenna to be detected.

[0171] Based on this, the use of drones to directly obtain the position and attitude information of the antenna reduces the need for manual measurement and improves the efficiency and accuracy of data acquisition. This method is particularly suitable for inaccessible areas or large-scale infrastructure monitoring. Secondly, by automatically processing the camera azimuth to determine the antenna orientation, human errors in data processing are reduced, ensuring the consistency and reliability of the results. In addition, this method provides a standardized antenna detection process that can be widely used in different drone systems, enhancing the versatility and adaptability of the technology.

[0172] Based on the above embodiment, in another embodiment provided by the present disclosure, the above step S530 may further include:

[0173] When the posture type is side-facing, the rotation angle of the rectangular detection frame is obtained, and the rotation angle is determined as the downtilt angle of the antenna to be detected.

[0174] In the embodiment, when the antenna is facing sideways, the side view of the antenna provides an important perspective on its downtilt angle. By obtaining the rotation angle of the rectangular detection frame when facing the side of the antenna, and determining this angle as the downtilt angle of the antenna to be detected, the downtilt angle of the antenna to be detected can be accurately evaluated.

[0175] If the antenna posture is other than the front or side facing, no processing will be performed.

[0176] Based on this, the side view of the antenna provided by the drone can intuitively capture the tilt of the antenna. Secondly, through image processing technology, the detection frame of the antenna is automatically identified and its rotation angle is calculated. This automated process greatly reduces the need for manual measurement and improves the speed of data collection and processing.

[0177] One or more technical solutions provided in the exemplary embodiments of the present disclosure control the drone to fly near the antenna, and photograph the front and side of each antenna one by one; perform antenna posture detection on the photographed photos. If the target antenna in the photo is detected as facing the front, the station height and longitude and latitude of the antenna can be obtained by reading the altitude and longitude and latitude of the drone. At the same time, read the azimuth of the camera at this time, and the direction angle of the antenna can be obtained after flipping 180°. If the target antenna in the photo is detected as facing the side, the downtilt angle of the target antenna is obtained by the rotation angle result of the rectangular detection frame. If the target antenna in the photo is detected at other angles, that is, not facing the front or the side, it will not be processed. After completing the station height, longitude and latitude, direction angle and downtilt angle of the target antenna, the measurement of the antenna engineering parameters is completed.

[0178] Among them, YOLOv5 is used to detect the antenna posture when facing forward and sideways. For the side-facing one, the rotation angle of the detection frame is also output. YOLOv5 optimizes the bounding box loss function and determines the rotation angle by calculating the IoU of the rotated rectangular box.

[0179] Therefore, the base station antenna engineering parameter measurement method provided in the exemplary embodiment of the present disclosure accurately matches the outer contour of the antenna by rotating the detection frame, thereby providing accurate antenna engineering parameter data.

[0180] The above mainly introduces the solutions provided by the exemplary embodiments of the present disclosure. It is understandable that in order to implement the above functions, the electronic device includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of each example described in the embodiments disclosed herein, the present disclosure can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present disclosure.

[0181] The exemplary embodiments of the present disclosure can divide the functional units of the electronic device according to the above method examples. For example, each functional module can be divided according to each function, or two or more functions can be integrated into one processing module. The above integrated modules can be implemented in the form of hardware or in the form of software functional modules. It should be noted that the division of modules in the exemplary embodiments of the present disclosure is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0182] In the case of dividing each functional module according to each function, an exemplary embodiment of the present disclosure provides a base station antenna engineering parameter measurement device, which can be a server or a chip applied to a server. Figure 6 This is a schematic block diagram of the functional modules of a base station antenna engineering parameter measurement device provided by an exemplary embodiment of the present disclosure. Figure 6 As shown, the base station antenna engineering parameter measurement device 600 includes:

[0183] The data acquisition module 610 is used to take a picture of the antenna to be tested by using a drone, and obtain the working parameters of the drone when taking the picture;

[0184] The data processing module 620 is used to input the image into a preset antenna posture detection model to obtain the posture type of the antenna to be detected and the rotation angle of the rectangular detection frame; the rectangular detection frame is used to locate the outer contour of the antenna to be detected;

[0185] The data processing module 620 is further used to determine the working parameter data of the antenna to be detected based on the working parameters of the drone, the posture type and the rotation angle of the rectangular detection frame.

[0186] In another embodiment provided by the present disclosure, the data processing module 620 further includes: the loss function of the antenna attitude detection model includes a bounding box loss function, the bounding box loss function is used to measure the closeness between the predicted rectangular detection box and the true bounding box, and the loss value of the bounding box loss function is determined based on the ratio of the intersection area and the union area between the rectangular detection box and the true bounding box.

[0187] In another embodiment provided by the present disclosure, the data processing module 620 is further used to respectively obtain the position information of the true bounding box and the rectangular detection box; respectively determine the area of ​​the true bounding box, the area of ​​the rectangular detection box and the area of ​​the intersection region based on the position information; determine the area of ​​the union region based on the area of ​​the true bounding box, the area of ​​the rectangular detection box and the area of ​​the intersection region; determine the loss value of the bounding box loss function based on the area of ​​the intersection region and the area of ​​the union region.

[0188] In another embodiment provided by the present disclosure, the data processing module 620 is further used to obtain the vertex position information, width, height and rotation angle of the real bounding box; determine the area of ​​the real bounding box based on the width and height of the real bounding box; obtain the vertex position information, width, height and rotation angle of the rectangular detection box; determine the area of ​​the rectangular detection box based on the width and height of the rectangular detection box; determine the area of ​​the intersection area based on the vertex position information and rotation angle of the real bounding box, and the vertex position information and rotation angle of the rectangular detection box.

[0189] In another embodiment provided by the present disclosure, the data processing module 620 is further used to detect the intersection of the real bounding box and the rectangular detection box based on the vertex position information and the rotation angle of the real bounding box, and the vertex position information and the rotation angle of the rectangular detection box, and obtain multiple vertex position information of the intersection area; sort the multiple vertex position information to form a closed polygon; divide the polygon into a plurality of triangles, and calculate the areas corresponding to the plurality of triangles respectively based on the plurality of vertex position information; accumulate the areas corresponding to the plurality of triangles respectively, and use the accumulated result as the area of ​​the intersection area.

[0190] In another embodiment provided by the present disclosure, the UAV operating parameters include the UAV altitude, longitude and latitude, and camera azimuth; the data processing module 620 is further used to determine the UAV altitude and longitude and latitude as the station altitude and longitude and latitude of the antenna to be detected; when the posture type is frontal, the camera azimuth is obtained, and the camera azimuth is flipped 180 degrees and determined as the azimuth of the antenna to be detected.

[0191] In another embodiment provided by the present disclosure, the data processing module 620 is further used to obtain the rotation angle of the rectangular detection frame when the posture type is side-facing, and determine the rotation angle as the downtilt angle of the antenna to be detected.

[0192] The exemplary embodiment of the present disclosure also provides an electronic device, comprising: at least one processor; and a memory connected to the at least one processor in communication. The memory stores a computer program that can be executed by the at least one processor, and the computer program is used to cause the electronic device to perform the method according to the embodiment of the present disclosure when executed by the at least one processor.

[0193] The exemplary embodiments of the present disclosure also provide a non-transitory computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor of a computer, is used to cause the computer to perform the method according to the embodiments of the present disclosure.

[0194] Figure 7 The block diagram of the electronic device provided as an example of the present disclosure, the block diagram of the electronic device 700 that can be used as the server or client of the present disclosure will now be described, which is an example of a hardware device that can be applied to various aspects of the present disclosure. The electronic device is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples, and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0195] like Figure 7 As shown, the electronic device 700 includes a computing unit 701, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 702 or a computer program loaded from a storage unit 708 into a random access memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the electronic device 700 can also be stored. The computing unit 701, the ROM 702, and the RAM 703 are connected to each other via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0196] A plurality of components in the electronic device 700 are connected to the I / O interface 705, including: an input unit 706, an output unit 707, a storage unit 708, and a communication unit 709. The input unit 706 may be any type of device capable of inputting information to the electronic device 700, and the input unit 706 may receive input digital or character information, and generate key signal inputs related to user settings and / or function control of the electronic device. The output unit 707 may be any type of device capable of presenting information, and may include but is not limited to a display, a speaker, a video / audio output terminal, a vibrator, and / or a printer. The storage unit 708 may include but is not limited to a disk, an optical disk. The communication unit 709 allows the electronic device 700 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks, and may include but is not limited to a modem, a network card, an infrared communication device, a wireless communication transceiver, and / or a chipset, such as a Bluetooth™ device, a WiFi device, a WiMax device, a cellular communication device, and / or the like.

[0197] The computing unit 701 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 701 performs the various methods and processes described above. Each of the methods described above may be implemented as a computer software program, which is tangibly included in a machine-readable medium, such as a storage unit 708. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 700 via the ROM 702 and / or the communication unit 709.

[0198] Figure 8 This is a schematic diagram of an exemplary computer program product provided by the present disclosure. The exemplary embodiment of the present disclosure also provides a computer program product 800, including a computer program 801, wherein the computer program 801 is used to enable the computer to perform a method according to an embodiment of the present disclosure when executed by a processor of the computer.

[0199] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0200] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0201] As used in this disclosure, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, apparatus, and / or device (e.g., disk, optical disk, memory, programmable logic device (PLD)) for providing machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term "machine-readable signal" refers to any signal for providing machine instructions and / or data to a programmable processor.

[0202] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0203] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0204] A computer system may include clients and servers. Clients and servers are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship to each other.

[0205] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instruction is loaded and executed on a computer, the process or function described in the embodiment of the present disclosure is executed in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, a terminal, a user device or other programmable device. The computer program or instruction may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer program or instruction may be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server, data center, etc. that integrates one or more available media. The available medium may be a magnetic medium, for example, a floppy disk, a hard disk, a tape; it may also be an optical medium, for example, a digital video disc (DVD); it may also be a semiconductor medium, for example, a solid state drive (SSD).

[0206] Although the present disclosure has been described in conjunction with specific features and embodiments thereof, it is apparent that various modifications and combinations may be made thereto without departing from the spirit and scope of the present disclosure. Accordingly, this specification and the drawings are merely exemplary illustrations of the present disclosure as defined by the appended claims, and are deemed to have covered any and all modifications, variations, combinations or equivalents within the scope of the present disclosure. Obviously, those skilled in the art may make various modifications and variations to the present disclosure without departing from the spirit and scope of the present disclosure. Thus, if these modifications and variations of the present disclosure fall within the scope of the claims of the present disclosure and their equivalents, the present disclosure is also intended to include these modifications and variations.

Claims

1. A method for measuring engineering parameters of a base station antenna, characterized in that: The method comprises: Taking pictures of the antenna to be tested by a drone, and obtaining the working parameters of the drone when taking the pictures; Input the image into a preset antenna posture detection model to obtain the posture type of the antenna to be detected and the rotation angle of a rectangular detection frame; the rectangular detection frame is used to locate the outer contour of the antenna to be detected; Based on the working parameters of the UAV, the posture type and the rotation angle of the rectangular detection frame, the working parameter data of the antenna to be detected is determined.

2. The method according to claim 1, characterized in that The loss function of the antenna attitude detection model includes a bounding box loss function, which is used to measure the proximity between the predicted rectangular detection box and the true bounding box. The loss value of the bounding box loss function is determined based on the ratio of the intersection area and the union area between the rectangular detection box and the true bounding box.

3. The method according to claim 2, characterized in that The loss value of the bounding box loss function is determined based on the ratio of the intersection area and the union area between the rectangular detection box and the true bounding box, including: Respectively obtain the position information of the real bounding box and the rectangular detection box; Based on the position information, respectively determine the area of ​​the real bounding box, the area of ​​the rectangular detection box, and the area of ​​the intersection region; Determine the area of ​​the union region based on the area of ​​the true bounding box, the area of ​​the rectangular detection box, and the area of ​​the intersection region; A loss value of the bounding box loss function is determined based on the area of ​​the intersection region and the area of ​​the union region.

4. The method according to claim 3, characterized in that The determining the real bounding box area, the rectangular detection box area and the intersection area based on the position information respectively includes: Obtaining vertex position information, width, height, and rotation angle of the real bounding box; Determine the area of ​​the real bounding box based on the width and height of the real bounding box; Obtaining vertex position information, width, height and rotation angle of the rectangular detection frame; Determine the area of ​​the rectangular detection frame based on the width and height of the rectangular detection frame; The area of ​​the intersection region is determined based on the vertex position information and the rotation angle of the real bounding box and the vertex position information and the rotation angle of the rectangular detection box.

5. The method according to claim 4, characterized in that The determining the area of ​​the intersection region based on the vertex position information and the rotation angle of the real bounding box and the vertex position information and the rotation angle of the rectangular detection box includes: Based on the vertex position information and rotation angle of the real bounding box and the vertex position information and rotation angle of the rectangular detection box, detect the intersection of the real bounding box and the rectangular detection box to obtain multiple vertex position information of the intersection area; Sorting the plurality of vertex position information so as to form a closed polygon; Divide the polygon into a plurality of triangles, and calculate the areas corresponding to the plurality of triangles respectively based on the plurality of vertex position information; The areas corresponding to the plurality of triangles are accumulated, and the accumulated result is used as the area of ​​the intersection region.

6. The method according to claim 1, characterized in that The UAV operating parameters include the UAV altitude, longitude and latitude, and camera azimuth; The determining of the working parameter data of the antenna to be detected based on the working parameters of the drone, the attitude type and the rotation angle of the rectangular detection frame includes: Determine the altitude and longitude and latitude of the drone as the station altitude and longitude and latitude of the antenna to be detected; When the posture type is front-facing, the camera azimuth is obtained, and the camera azimuth is flipped 180 degrees to determine it as the azimuth of the antenna to be detected.

7. The method according to claim 6, characterized in that The method further comprises: When the posture type is side-facing, the rotation angle of the rectangular detection frame is obtained, and the rotation angle is determined as the downtilt angle of the antenna to be detected.

8. A computer device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the method according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images