Air taper sleeve target detection and positioning method

Abstract

The invention discloses an air taper sleeve target detection and positioning method, and belongs to the technical field of target tracking and positioning. The invention provides an air taper sleeve target tracking and positioning method. Firstly, near-infrared LED lamp beads are installed on a taper sleeve, and an infrared optical filter is installed on a camera lens, so that infrared target image acquisition is completed. Then, an improved YOLOv11 detection algorithm is provided, the small target detection capability of YOLO is improved, and infrared target tracking is realized by adopting a Byterack strategy. And finally, using an improved pose estimation method to improve the positioning precision of the taper sleeve position.

Description

Technical Field

[0001] The present invention relates to the technical field of target tracking and positioning, and in particular to an aerial drogue target detection and positioning method. Background Art

[0002] Moving target tracking and positioning has always been a challenging task and is also a hot research topic.

[0003] In terms of target tracking, early target tracking algorithms were mainly based on target modeling or tracking target features. The main methods are: 1) Method based on target modeling: By modeling the target appearance model, the target is found in the subsequent frames.

[0004] 2) Search-based methods. As research deepens, people find that methods based on target modeling process the entire image, which has poor real-time performance.

[0005] With the recent widespread adoption of deep learning methods, researchers have begun considering their application to object tracking. Currently, deep learning applications in object tracking can be categorized into two types: one that combines deep learning with correlation filtering, using deep learning as a feature extraction method and correlation filtering as a tracking framework; and another that relies solely on neural network frameworks. Among these, the YOLO series of algorithms has garnered significant attention. While these algorithms excel in many areas, such as real-time performance, they suffer from missed detections and false detections when detecting small objects. YOLO (You Only Look Once) is an end-to-end object detection algorithm.

[0006] In terms of pose positioning, traditional target positioning technologies primarily utilize sensors such as radar, laser, and infrared for distance measurement. However, these sensors have the following drawbacks: First, they are easily affected by other sensors on the machine, resulting in inaccurate ranging. Second, their greatest drawback is their inability to identify the target object. Visual positioning methods can be categorized as monocular vision-based positioning, binocular vision-based positioning, and omnidirectional vision-based positioning. Monocular vision-based positioning methods can be further categorized based on the number of images used for positioning: single-image, two-image, or more. Single-image-based positioning uses the description of known target features (such as points, lines, curves, and plane quadrilaterals) in the world coordinate system, the projection of these features onto the image, and some additional feature parameters to determine the pose relationship between the world coordinate system and the camera coordinate system to locate the target. Summary of the Invention

[0007] The present invention aims to address these shortcomings by providing a method for detecting and locating aerial drogue targets. By installing near-infrared LEDs on the drogue and an infrared filter on the camera lens, infrared target image acquisition is accomplished. An improved YOLOv11 detection algorithm is then proposed to enhance YOLO's ability to detect small targets, and a Bytetrack strategy is employed to achieve infrared target tracking. Finally, an improved pose estimation method is employed to improve the accuracy of drogue positioning.

[0008] The present invention is achieved through the following solutions: A method for detecting and locating an aerial drogue target comprises the following steps: Step S1, model training, using the near-infrared annotated dataset of the aerial drogue to train the YOLO detection model; Step S2, calibrating the camera, solving the camera's intrinsic and extrinsic parameter matrix and distortion parameters, and optimizing the camera's intrinsic parameters using the CPSO algorithm; Step S3, performing image preprocessing on the real-time collected images using homomorphic filtering technology; Step S4, using the trained YOLO detection model and Bytetrack strategy to complete real-time target detection and tracking of the aerial drogue; Step S5: extract the region of interest based on the bounding box detected by the Yolo detection model, filter it, binarize it, and convert it into a binary image; Step S6, performing Hough transform on the spot area of ​​the binary image, fitting the ellipse formed by the spot, solving the coordinates of the center of the ellipse to obtain the coordinates of the center position of the cone sleeve on the image; In step S7, the relative position of the target and the camera in the world coordinate system is calculated using the calibrated camera parameters. Then, based on the position offset between the camera and a fixed point of the drone, the position offset between a fixed point of the drone and the center of the drogue is calculated.

[0009] In step S1, the specific steps of infrared target image acquisition are: The eight infrared LED light sources installed on the cone sleeve serve as target identification objects. The eight infrared LED light sources are evenly spaced around the circumference of the cone sleeve. The camera lens is equipped with an infrared filter, and filtered sampling is performed through the camera.

[0010] In step S2, the camera is calibrated using the Zhang Zhengyou calibration method and the camera internal parameters are optimized; specifically, Construct a mathematical model based on the imaging relationship, the linear relationship between the image coordinate system, the pixel coordinate system, the camera coordinate system and the world coordinate system; The chaotic particle swarm optimization algorithm is used in the optimization process of internal and external parameters of camera calibration, so that the camera parameter solution converges to the global optimal solution with a high probability, and obtains calibration parameters with higher accuracy; the chaotic particle swarm optimization algorithm uses the Logistics equation as the chaotic system.

[0011] In step S2, the mathematical model constructed is:

[0012] Where: u, v are the horizontal and vertical coordinates of the image coordinate system, S is the distance from the camera image plane to the object along the optical axis, dx, dy are the horizontal and vertical dimensions of the pixel, u0, v0 are the center positions of the image plane, f is the distance to the camera, R is the rotation matrix of the calibration object, t is the translation matrix, kx, ky are the magnification factors from the imaging coordinates to the image coordinates on the focal length normalized imaging plane, respectively, and X, Y and Z are the positions of the feature points in the world coordinate system; The following formula is obtained by simplifying the mathematical model:

[0013] Where: S is the scaling factor scalar, M is the object plane coordinate, A1 is the camera's intrinsic parameter matrix, A2 is the camera's extrinsic parameter matrix, and H is the homography matrix, which contains both the camera's intrinsic and extrinsic parameters. The required feature points are determined through the formula, and the chaotic particle swarm optimization algorithm is used in the process of optimizing the internal and external parameters of the camera calibration; Specifically: Use the CPSO algorithm to optimize the following objective function:

[0014] Get the optimal solution of the camera's internal parameters, where: The world coordinate point M is projected through the camera imaging model; m ij is the pixel coordinate point obtained by corner point detection: The evolutionary strategy of inertia weight is adopted to reduce nonlinearity. The specific formula is: : In the formula, iter max is the maximum number of iterations, w k is the inertia weight at the kth iteration, w min 、w max w k The value range of .

[0015] Step S3 is specifically as follows: before target detection and tracking, the image will be preprocessed using homomorphic filtering technology. Homomorphic filtering is a technology used for image enhancement and restoration. It separates the illumination component and the reflection component of the image, processes them separately, and then recombine them. The specific steps include: logarithmic transformation, Fourier transform, high-pass filtering, inverse Fourier transform, and exponential transformation.

[0016] In step S4, the trained YOLO detection model and Bytetrack strategy are used to complete the real-time target detection and tracking of the aerial drogue. Specifically, the following steps are included: Step S41: Improve the YOLO algorithm by replacing all ReLu activation functions in the YOLOv11 backbone network with SiLU functions; introduce the ECANet module into the YOLOv11 backbone network; Step S42: Target tracking based on yolo-Bytetrack. After completing the target detection of the cone sleeve, the YOLOv11 detector will use Bytetrack to complete the tracking. Bytetrack is a tracking algorithm based on target detection. It uses Kalman filtering to predict the bounding box and uses the Hungarian algorithm to match the target detection box and the prediction box. In addition, two associations are performed through the high-confidence detection box and the low-confidence detection box.

[0017] In step S41, for a given input feature, ECANet first performs a global average pooling operation channel by channel while keeping the channel dimension unchanged; then, the processed data is passed into a 1×1 convolution for information fusion; subsequently, the data is activated by a Sigmoid function and multiplied with the original feature to fully fuse the global features.

[0018] In step S5, it specifically includes the following steps: Step S51: Extracting the region of interest. Assume that the bounding box position detected by Yolo is (x1, y2, h, w), where x, y, h, and w are the center horizontal coordinate, center horizontal coordinate, height, and width of the detection box, respectively. When cropping, to ensure that the detection target is within the cropping area, the bounding box will be expanded outward by a certain width d. The bounding box information of the region of interest is (x1, y2, h+d, w+d); Step S52: Calculate the center coordinates of the front ellipse of the cone sleeve, and the algorithm performs filtering and binarization processing on the cropped region of interest.

[0019] In step S6, it specifically includes the following steps: Step S61, solve the center coordinates of the ellipse: Use the Hough transform to fit the light spot in the area to an ellipse, and solve for the center coordinates of the ellipse in the o'-x'y' coordinate system (x2, y2). The detection bounding box information in the O-xy coordinate system is (x1, y2, h, w). The coordinates of point o' in the O-xy coordinate system are (x1-0.5hd, y1-0.5hd). The center coordinates of the ellipse in the O-xy coordinate system are solved as (x1+x2-0.5hd, y1+y2-0.5hd). Step S62: Calculate the coordinates of the center of the cone sleeve on the image, the linear relationship between the image coordinate system, the pixel coordinate system, the camera coordinate system, and the world coordinate system, and calibrate the solved camera internal and external parameters and the homography matrix H value. Use the relevant formula to calculate the coordinates of the cone sleeve center relative to the camera in the real world: .

[0020] In step S7, it specifically includes the following steps: Since there is a coordinate deviation between the camera and the fixed point of the drone in the real world (Δ x , Δ y , Δ z ), specifically, the deviation value is obtained through real-world measurement. The coordinate deviation must be considered when solving the real-world relative position of the drogue and a fixed point of the drone. The coordinate of the drogue relative to the drone is W +(Δ x , Δ y , Δ z ).

[0021] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: This solution provides an improved YOLOv11 detection algorithm to enhance YOLO's ability to detect small infrared targets. The tracking algorithm uses the Bytetrack strategy. An improved pose estimation method is then used to measure the target position, improving the target's positioning accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 The following is a technical flow chart of a method for tracking and locating an aerial drogue target; Figure 2 Search flow chart when optimizing camera internal parameters for CPSO algorithm; Figure 3 This is a plane effect diagram of the infrared LED installed on the front of the cone sleeve proposed by the present invention; Figure 4 This is the 3D rendering of the cone sleeve; Figure 5 This is the front effect picture of the cone sleeve with infrared LED taken by the camera under infrared imaging; Figure 6 Improved diagram for the YOLOv11 backbone network; Figure 7 This is the flow chart of the yolo-Bytetrack target tracking algorithm; Figure 8 Extract the method map for the region of interest; Figure 9 This is the ellipse fitting effect diagram; Figure 10 is the region of interest-camera coordinate transformation diagram. DETAILED DESCRIPTION

[0023] All features disclosed in this specification, or all steps in the disclosed methods or processes, except mutually exclusive features and / or steps, can be combined in any manner.

[0024] Any feature disclosed in this specification (including any appended claims and abstract), unless otherwise stated, may be replaced by other equivalent or similar features. In other words, unless otherwise stated, each feature is only an example of a series of equivalent or similar features.

[0025] In the description of the present invention, it should be understood that the terms "up", "down", "left", "right", etc., indicating directions or positional relationships, 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 element referred to must have a predetermined direction, be constructed and operated in a predetermined direction, and therefore cannot be understood as a limitation on the present invention.

[0026] Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be understood to indicate or imply relative importance or to implicitly indicate the quantity of the technical features being referred to. Thus, a feature defined as "first," "second," etc. may explicitly or implicitly include one or more of such features.

[0027] Example 1 The present invention provides a technical solution: like Figure 1 As shown, a method for detecting and locating an aerial drogue target includes the following steps: Step S1, model training, using the near-infrared annotated dataset of the aerial drogue to train the YOLO detection model; In step S2, the camera is calibrated using the Zhang Zhengyou calibration method to solve the camera's intrinsic and extrinsic parameter matrix and distortion parameters, and the camera's intrinsic parameters are optimized using the CPSO algorithm (chaotic particle swarm optimization algorithm).

[0028] Step S3, using homomorphic filtering technology to perform image preprocessing on the real-time collected image to solve the problem of uneven illumination of the image; Step S4, using the trained YOLO detection model and Bytetrack strategy to complete real-time target detection and tracking of the aerial drogue; Step S5: extract the region of interest (ROI) based on the bounding box detected by the Yolo detection model, filter and binarize it, and convert it into a binary image; Step S6, performing Hough transform on the spot area of ​​the binary image, fitting the ellipse formed by the spot, solving the coordinates of the center of the ellipse to obtain the coordinates of the center position of the cone sleeve on the image; In step S7, the relative position of the target and the camera in the world coordinate system is calculated using the calibrated camera parameters. Then, based on the position offset between the camera and a fixed point of the drone, the position offset between a fixed point of the drone and the center of the drogue is calculated.

[0029] In step S1, before model training, it is necessary to first prepare an infrared target dataset. The specific steps are as follows: The 8 infrared LED light sources installed on the cone sleeve are used as target identification objects. The 8 infrared LED light sources are evenly spaced around the circumference of the cone sleeve. Its front plane structure is as follows Figure 3 As shown, the 3D structure is Figure 4 As shown; the camera lens is equipped with an infrared filter, and its sampling effect is shown in the figure Figure 5 As shown; the YOLO model is based on Figure 5 The sample data set is used to train the detection model; In step S2, the present invention adopts the Zhang Zhengyou calibration method when calibrating the camera, and optimizes the camera internal parameters, specifically including the following steps: According to the imaging relationship, the linear relationship between the image coordinate system, pixel coordinate system, camera coordinate system and world coordinate system, the following mathematical model is constructed: (1) Where: u, v are the horizontal and vertical coordinates of the image coordinate system, S is the distance from the camera image plane to the object along the optical axis, dx, dy are the horizontal and vertical dimensions of the pixel, u0, v0 are the center positions of the image plane, f is the distance to the camera, R is the rotation matrix of the calibration object, t is the translation matrix, kx, ky are the magnification factors from the imaging coordinates to the image coordinates on the focal length normalized imaging plane, respectively, and X, Y and Z are the positions of the feature points in the world coordinate system; Simplifying formula (1) we can get: (2) Where: S is the scaling factor scalar, M is the object plane coordinate, A1 is the camera's intrinsic parameter matrix, A2 is the camera's extrinsic parameter matrix (the rotation and translation parameters of the calibration plate, [r1, r2, t]), and H is the homography matrix, which contains both the camera's intrinsic and extrinsic parameters.

[0030] From formula (2), we can see that the matrix equation can be transformed into two linear matrices. Therefore, in order to obtain the 9 unknown factors in the homography matrix H, at least 5 spatially known feature points are required. As the number of feature points increases, the LM method can obtain more accurate intrinsic parameter values. (The Levenberg-Marquardt method (LM method) is a nonlinear optimization algorithm that combines the gradient descent method and the Gauss-Newton method. It is mainly used to solve the least squares problem and is widely used in machine learning and engineering fields.)

[0031] In actual application, due to factors such as manufacturing and installation, the camera is bound to have various distortions. In order to describe the law of distortion imaging, a nonlinear imaging model is introduced. Therefore, there are deviations in actual lens imaging. These deviations need to be considered in the process of solving the internal and external parameters. Suppose the coordinates of the actual image point are (u d , v d ), the ideal coordinate is (u u , v u ), usually only considering radial distortion can also achieve higher accuracy and meet application requirements. The corresponding relationship is shown in formula (3): (3) Where: k1, k2, k3 are distortion parameters, representing the 2nd, 4th, and 6th order distortion parameters in the horizontal and vertical directions.

[0032] In the aforementioned Zhang Zhengyou camera calibration method, the camera's initial values ​​are obtained by assuming a linear model. These initial values ​​are then optimized using the LM method to solve for the initial value of the homography matrix H. However, Zhang Zhengyou's calibration method estimates distortion parameters based on the camera's linear model, and traditional optimization strategies like the LM method are prone to falling into local optimality, which reduces calibration accuracy. To improve the accuracy of camera calibration parameters, this solution uses a chaotic particle swarm optimization algorithm (CPSO) to optimize the camera's internal and external parameters during calibration. This ensures that the camera parameter solution converges to the global optimal solution with a high probability, resulting in more accurate calibration parameters. The CPSO algorithm is used to optimize the following objective function: (4) Get the optimal solution of the camera's internal parameters, where: The world coordinate point M is projected through the camera imaging model; m ij is the pixel coordinate point obtained by corner detection.

[0033] CPSO maps chaotic variables from chaotic space to solution space and then performs random search on the chaotic variables. Chaotic optimization algorithms are characterized by being insensitive to initial values, easy to escape local minima, fast search speed, high computational accuracy, and global asymptotic convergence.

[0034] The evolution strategy of inertia weight, the present invention adopts nonlinear reduction as shown in formula (5): (5) In the formula, iter max is the maximum number of iterations, w k is the inertia weight at the kth iteration, w min 、w max w k The value range of In addition, the CPSO algorithm uses the Logistics equation as the chaotic system. Compared with other chaotic systems, this equation has a smaller computational complexity, so this equation is used to construct the chaotic sequence. The specific search process of the Chaos Particle Swarm Optimization Algorithm (CPOS) is as follows: Figure 2 shown.

[0035] In step S3, the real-time acquired image is preprocessed using homomorphic filtering technology. Specifically, before target detection and tracking, the original image is preprocessed using homomorphic filtering technology. Homomorphic filtering is a technology used for image enhancement and restoration. It separates the illumination component and the reflection component of the image, processes them separately, and then recombines them. The specific steps include: logarithmic transformation, Fourier transform, high-pass filtering, inverse Fourier transform, and exponential transformation. This technology is effective in dealing with uneven illumination problems in images. In step S4, the trained YOLO detection model and Bytetrack strategy are used to complete the real-time target detection and tracking of the aerial drogue. Specifically, the following steps are included: Step S41: Improvement of the YOLO algorithm. To identify and track the cone sleeve, several training data sets with bounding box labels are required. The present invention takes into account that the cone sleeve is far away from the camera at high altitude, and the cone sleeve accounts for a small proportion of the sampled image, which may cause small target recognition. Therefore, the present invention proposes an improved detection and tracking algorithm based on YOLOv11. This algorithm replaces all ReLu activation functions in the YOLOv11 backbone network with SiLU functions. The SiLU function can effectively perform nonlinear transformations on the input of the neural network and is differentiable within the entire domain. The SiLU function is not monotonically increasing, but reaches a global minimum of −0.28 at x≈−1.28, which can act as an implicit regularization to suppress excessive weights.

[0036] Secondly, the algorithm also incorporates an attention mechanism module, the application of which can significantly improve target detection performance. Existing attention mechanisms include CBAM, SOCA, SENet, ECANet, etc. The addition of these different attention mechanisms has been proven in related studies to significantly improve the performance of the original YOLO network structure; ECANet is very suitable for CNN, and while effectively enhancing cross-channel interaction information, it can effectively avoid model complexity and information loss problems; in order to improve the model's detection performance for small targets, this paper introduces the ECANet module into the backbone network of YOLOv11, as shown below: Figure 6 shown.

[0037] For a given input feature, ECANet first performs a global average pooling operation on each channel while keeping the channel dimension unchanged. Then, the processed data is passed to a 1×1 convolution for information fusion. Subsequently, the data is activated by a Sigmoid function and multiplied with the original feature to fully fuse the global features. Step S42: Target tracking based on yolo-Bytetrack. After completing the target detection of the cone sleeve, the YOLOv11 detector will use Bytetrack to complete the tracking. Bytetrack is a tracking algorithm based on target detection. It uses Kalman filtering to predict the bounding box and uses the Hungarian algorithm to match the target detection box and the prediction box. In addition, by performing two associations between the high-confidence detection box and the low-confidence detection box, a good tracking effect can be achieved. The target tracking algorithm process based on yolo-Bytetrack is as follows: Figure 7 shown.

[0038] In step S5, the main idea of ​​this step is: to more accurately solve the position of the center of the cone sleeve and the camera, after completing the target bounding box detection, the algorithm will extract and crop the region of interest, and then use the Hough transform to fit the light spot in the cropped region of interest to find the center coordinates of the ellipse. This step specifically includes the following steps: Step S51: Extraction of the region of interest. Assume that the bounding box position detected by yolo is (x1, y2, h, w), where x, y, h, and w are the center horizontal coordinate, center horizontal coordinate, height, and width of the detection box, respectively. When cropping, in order to ensure that the detection target is within the cropping area, the bounding box will be expanded outward to a certain width d, such as Figure 8 As shown, the bounding box information of the region of interest is (x1, y2, h+d, w+d); Step S52: Calculate the center coordinates of the front ellipse of the cone sleeve, and the algorithm performs filtering and binarization processing on the cropped region of interest.

[0039] In step S6, the following steps are specifically included: Step S61, solve the center coordinates of the ellipse: Use the Hough transform to fit the light spot in the area to an ellipse, and solve for the center coordinates of the ellipse in the o'-x'y' coordinate system (x2, y2). As previously known, the detection bounding box information in the O-xy coordinate system is (x1, y2, h, w), so the coordinates of point o' in the O-xy coordinate system are (x1-0.5hd, y1-0.5hd); therefore, the center coordinates of the ellipse in the O-xy coordinate system (image coordinate system) are solved as (x1+x2-0.5hd, y1+y2-0.5hd); In step S62, the coordinates of the center of the cone sleeve on the image are solved. Based on the linear relationship between the image coordinate system, pixel coordinate system, camera coordinate system and world coordinate system, as well as the calibrated camera internal and external parameters and the homography matrix H value, the coordinates of the cone sleeve center relative to the camera in the real world can be calculated using formula (2): (7) In step S7, specifically, due to the coordinate deviation (Δ x , Δ y , Δ z ), the specific deviation value can be obtained through real-world measurement. Therefore, the coordinate deviation must be considered when solving the real-world relative position of the drogue and a fixed point of the drone. The coordinate of the drogue relative to the drone is W +(Δ x , Δ y , Δ z ).

[0040] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for detecting and locating an aerial drogue target, characterized by: The following steps are involved: Step S1, infrared target image acquisition, and use the near-infrared annotated dataset of the aerial drogue to train the YOLO detection model; Step S2, calibrating the camera, solving the camera's intrinsic and extrinsic parameter matrix and distortion parameters, and optimizing the camera's intrinsic parameters using the CPSO algorithm; Step S3, performing image preprocessing on the real-time collected images using homomorphic filtering technology; Step S4, using the trained YOLO detection model and Bytetrack strategy to complete real-time target detection and tracking of the aerial drogue; Step S5: extract the region of interest based on the bounding box detected by the Yolo detection model, filter it, binarize it, and convert it into a binary image; Step S6, performing Hough transform on the spot area of ​​the binary image, fitting the ellipse formed by the spot, solving the coordinates of the center of the ellipse to obtain the coordinates of the center position of the cone sleeve on the image; In step S7, the relative position of the target and the camera in the world coordinate system is calculated using the calibrated camera parameters. Then, based on the position offset between the camera and a fixed point of the drone, the position offset between a fixed point of the drone and the center of the drogue is calculated.

2. The method for detecting and locating an aerial drogue target according to claim 1, wherein: In step S1, the specific steps of infrared target image acquisition are: The eight infrared LED light sources installed on the cone sleeve serve as target identification objects. The eight infrared LED light sources are evenly spaced around the circumference of the cone sleeve. The camera lens is equipped with an infrared filter, and filtered sampling is performed through the camera.

3. The method for detecting and locating an aerial drogue target according to claim 2, wherein: In step S2, the camera is calibrated using the Zhang Zhengyou calibration method and the camera internal parameters are optimized; specifically, Construct a mathematical model based on the imaging relationship, the linear relationship between the image coordinate system, the pixel coordinate system, the camera coordinate system and the world coordinate system; The chaotic particle swarm optimization algorithm is used in the optimization process of internal and external parameters of camera calibration, so that the camera parameter solution converges to the global optimal solution with a high probability, and obtains calibration parameters with higher accuracy; the chaotic particle swarm optimization algorithm uses the Logistics equation as the chaotic system.

4. The method for detecting and locating an aerial drogue target according to claim 3, wherein: In step S2, the mathematical model constructed is: Where: u, v are the horizontal and vertical coordinates of the image coordinate system, S is the distance from the camera image plane to the object along the optical axis, dx, dy are the horizontal and vertical dimensions of the pixel, u0, v0 are the center positions of the image plane, f is the distance to the camera, R is the rotation matrix of the calibration object, t is the translation matrix, kx, ky are the magnification factors from the imaging coordinates to the image coordinates on the focal length normalized imaging plane, respectively, and X, Y and Z are the positions of the feature points in the world coordinate system; The following formula is obtained by simplifying the mathematical model: Where: S is the scaling factor scalar, M is the object plane coordinate, A1 is the camera's intrinsic parameter matrix, A2 is the camera's extrinsic parameter matrix, and H is the homography matrix, which contains both the camera's intrinsic and extrinsic parameters. The required feature points are determined through the formula, and the chaotic particle swarm optimization algorithm is used in the process of optimizing the internal and external parameters of the camera calibration; Specifically: Use the CPSO algorithm to optimize the following objective function: Get the optimal solution of the camera's internal parameters, where: The world coordinate point M is projected through the camera imaging model; m ij is the pixel coordinate point obtained by corner point detection: The evolutionary strategy of inertia weight is adopted to reduce nonlinearity. The specific formula is: ： In the formula, iter max is the maximum number of iterations, w k is the inertia weight at the kth iteration, w min 、w max w k The value range of .

5. The method for detecting and locating an aerial drogue target according to claim 4, wherein: Step S3 is specifically as follows: before target detection and tracking, the image will be preprocessed using homomorphic filtering technology. Homomorphic filtering is a technology used for image enhancement and restoration. It separates the illumination component and the reflection component of the image, processes them separately, and then recombine them. The specific steps include: logarithmic transformation, Fourier transform, high-pass filtering, inverse Fourier transform, and exponential transformation.

6. The method for detecting and locating an aerial drogue target according to claim 5, wherein: In step S4, the trained YOLO detection model and Bytetrack strategy are used to complete the real-time target detection and tracking of the aerial drogue. Specifically, the following steps are included: Step S41: Improve the YOLO algorithm by replacing all ReLu activation functions in the YOLOv11 backbone network with SiLU functions; introduce the ECANet module into the YOLOv11 backbone network; Step S42: Target tracking based on yolo-Bytetrack. After completing the target detection of the cone sleeve, the YOLOv11 detector will use Bytetrack to complete the tracking. Bytetrack is a tracking algorithm based on target detection. It uses Kalman filtering to predict the bounding box and uses the Hungarian algorithm to match the target detection box and the prediction box. In addition, two associations are performed through the high-confidence detection box and the low-confidence detection box.

7. The method for detecting and locating an aerial drogue target according to claim 6, wherein: In step S41, for a given input feature, ECANet first performs a global average pooling operation channel by channel while keeping the channel dimension unchanged; then, the processed data is passed into a 1×1 convolution for information fusion; subsequently, the data is activated by a Sigmoid function and multiplied with the original feature to fully fuse the global features.

8. The method for detecting and locating an aerial drogue target according to claim 7, wherein: In step S5, it specifically includes the following steps: Step S51: Extracting the region of interest. Assume that the bounding box position detected by Yolo is (x1, y2, h, w), where x, y, h, and w are the center horizontal coordinate, center horizontal coordinate, height, and width of the detection box, respectively. When cropping, to ensure that the detection target is within the cropping area, the bounding box will be expanded outward by a certain width d. The bounding box information of the region of interest is (x1, y2, h+d, w+d); Step S52: Calculate the center coordinates of the front ellipse of the cone sleeve, and the algorithm performs filtering and binarization processing on the cropped region of interest.

9. The method for detecting and locating an aerial drogue target according to claim 8, wherein: In step S6, it specifically includes the following steps: Step S61, solve the center coordinates of the ellipse: Use the Hough transform to fit the light spot in the area to an ellipse, and solve for the center coordinates of the ellipse in the o'-x'y' coordinate system (x2, y2). The detection bounding box information in the O-xy coordinate system is (x1, y2, h, w). The coordinates of point o' in the O-xy coordinate system are (x1-0.5hd, y1-0.5hd). The center coordinates of the ellipse in the O-xy coordinate system are solved as (x1+x2-0.5hd, y1+y2-0.5hd). Step S62: Calculate the coordinates of the center of the cone sleeve on the image, the linear relationship between the image coordinate system, the pixel coordinate system, the camera coordinate system, and the world coordinate system, and calibrate the solved camera internal and external parameters and the homography matrix H value. Use the relevant formula to calculate the coordinates of the cone sleeve center relative to the camera in the real world: 。 10. The aerial drogue target detection and positioning method according to claim 9, characterized in that: In step S7, it specifically includes the following steps: Since there is a coordinate deviation between the camera and the fixed point of the drone in the real world (Δ x , Δ y , Δ z ), specifically, the deviation value is obtained through real-world measurement. The coordinate deviation must be considered when solving the real-world relative position of the drogue and a fixed point of the drone. The coordinate of the drogue relative to the drone is W +(Δ x , Δ y , Δ z ).

Images