A method and apparatus for adjusting a shipborne water cannon platform based on jet deviation prediction and ship attitude prediction

Abstract

This invention provides a method and apparatus for adjusting a shipborne water cannon platform based on jet deviation prediction and ship attitude prediction, belonging to the field of water cannon platform adjustment technology. The method includes: acquiring real-time video frame images, real-time brightness values, real-time wave height, real-time wave frequency, real-time air humidity, and real-time visibility; establishing an image segmentation model incorporating brightness values, wave height, wave frequency, air humidity, and visibility, and inputting the real-time data into the image segmentation model to segment the water column image, the impact point image, and the target image; determining the actual distance deviation and iteratively adjusting the yaw and pitch angles of the water cannon platform until the target is completely hit. In image segmentation, this invention uses an image segmentation model incorporating brightness values, wave height, wave frequency, air humidity, and visibility to segment the image, fully considering the influence of external environmental factors, resulting in more accurate segmented images and ensuring precise adjustment of the water cannon platform to hit the target.

Description

Technical Field

[0001] This invention relates to the field of water cannon platform adjustment technology, specifically to a method and apparatus for adjusting a shipborne water cannon platform based on jet deviation prediction and ship attitude prediction. Background Technology

[0002] With the development of global economic globalization, more ships are traveling on major shipping routes around the world. However, issues such as haphazard anchoring obstructing traffic, foreign vessels crossing borders, and piracy robbing merchant ships are disrupting the normal development of the global maritime industry. A common method to address these problems is to use water cannons for dispersal; however, these enforcement water cannons require manual observation to determine if a jet is about to strike, to adjust the cannon's caliber and orientation, and to observe whether it has hit a pirate ship—a tedious and physically demanding task.

[0003] In the prior art, a method for evaluating the ship's impact effect based on jet edge analysis and impact deviation feedback, disclosed in CN110245569B, includes the following steps: S1 Acquire video image information through an optoelectronic pan-tilt device; S2 Process the video image information using a jet segmentation model to extract the jet contour and ship contour; S3 Determine whether a jet impact exists based on the extraction results. If no jet is found, return to S2; if a jet is found, proceed to S4; S4 Determine whether the jet hits the ship based on contour analysis. If the jet does not hit the ship, proceed to S5; if the jet hits the ship, use the ship's position as the impact coordinates and proceed to S7; S5 Calculate the deviation in water cannon firing by combining the target ship's coordinates with the actual impact point of the jet; S6 Convert the actual impact point of the jet and the calculated deviation into the water cannon's impact coordinates using a coordinate transformation method; S7 Feedback the impact coordinates to the water cannon controller to correct the water cannon's control parameters. This method can not only determine whether a jet attack has occurred, but also obtain the actual point of impact, feeding back to the water cannon system to adjust the cannon's firing direction and achieve precise strikes against pirate ships. It solves the problems of determining whether a target has been hit and obtaining the desired strike effect that are difficult to achieve with traditional methods.

[0004] However, existing technologies still have significant shortcomings. For example, when processing video images to extract jet and ship contours, existing technologies do not consider the impact of external environmental factors on image segmentation accuracy (such as the brightness of the target area, air humidity, visibility, and water surface disturbances, all of which affect image segmentation accuracy). This results in inaccurate jet and ship contours formed by segmentation, which in turn adversely affects subsequent calculation deviations and the process of adjusting the water cannon angle, making it difficult to accurately hit the target even after the water cannon is adjusted.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a method and apparatus for adjusting a shipborne water cannon platform based on jet deviation prediction and ship attitude prediction, so as to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A method for adjusting a shipborne water cannon platform based on jet deviation prediction and ship attitude prediction, comprising the following steps:

[0009] S1, record the water cannon strike area in real time to obtain real-time video frame images of the water cannon strike area, process the real-time video frame images to obtain the real-time brightness value, real-time wave height and real-time wave frequency of the water cannon strike area, and obtain the real-time air humidity and real-time visibility at the water cannon platform.

[0010] S2, establish an image segmentation model that incorporates brightness value, wave height, wave frequency, air humidity and visibility, and input the data collected in step S1 into the image segmentation model to segment the water column image, the impact point image and the target image to be hit;

[0011] S3, calculate the cross-union ratio between the impact point image and the target image. If the cross-union ratio is greater than the cross-union ratio threshold, continue firing at the current muzzle angle. If the cross-union ratio is not greater than the cross-union ratio threshold, proceed to step S4.

[0012] S4. Based on target tracking technology, the three-dimensional coordinates of the target to be hit are obtained, and the three-dimensional coordinates of the target to be hit and the impact point image are processed to generate the physical distance deviation. Based on ship attitude prediction technology, the water cannon distance deviation between the water cannon and the current water cannon at the next water cannon strike is predicted. The physical distance deviation and the water cannon distance deviation are combined to generate the comprehensive distance deviation.

[0013] S5, based on the comprehensive distance deviation, iteratively adjusts the yaw and pitch angles of the water cannon platform until the crossover ratio is greater than the crossover ratio threshold.

[0014] Furthermore, step S2 includes the following steps;

[0015] S21. An image segmentation model is established based on the deeplabv3+ model. The lightweight network MobileNetV2 is selected as the backbone network, and the CBAM attention module and data preprocessing module are inserted into the image segmentation model.

[0016] S22, Obtain the historical sample set, which includes historical video frame images, historical brightness values, historical wave heights, historical wave frequencies, historical air temperatures, and historical visibility. The historical sample set is randomly divided into a training set and a test set. Water columns, impact points, and targets to be hit are marked on the historical video frame images in the training set.

[0017] S23, input the historical sample set into the preprocessing module of the image segmentation model for preprocessing. The preprocessing includes denoising the video frame image and normalizing the video frame image, brightness value, wave height, wave frequency, air temperature and visibility.

[0018] S24, the image segmentation model is trained based on the preprocessed training set, and the trained image segmentation model is tested based on the preprocessed test set until a qualified image segmentation model is obtained. The real-time data collected in step S1 is input into the qualified image segmentation model to segment the water column image, the impact point image, and the target image to be hit.

[0019] Furthermore, the specific logic for training the image segmentation model using the preprocessed training set is as follows:

[0020] 1.1, Feature extraction is performed on video frame images based on depthwise separable convolution technology to generate a first feature map;

[0021] 1.2 Based on the preprocessed brightness value, wave height, wave frequency, air humidity, and visibility, an environmental impact value is generated. The calculation formula is as follows:

[0022]

[0023] Where E is the environmental impact value, L, V, RH, H, f are the pre-processed brightness value, visibility, air humidity, wave height and wave frequency respectively, V0 and RH0 are the visibility threshold and air humidity threshold respectively, λ1, λ2, λ3, λ4, C1, γ, δ, ω1, ω2 are all undetermined coefficients, and ω1+ω2=1.

[0024] 1.3. The first feature map and the environmental impact value are concatenated to form a comprehensive feature map, and the comprehensive feature map is then subjected to Atrous convolution to obtain the second feature map. The formula used is as follows:

[0025]

[0026] Where I1 represents the first feature map, Concat(I1,E) concatenates the first feature map and the environmental influence value, Iz represents the comprehensive feature map, Iz(x+r*k',y) represents the pixel value at position (x+r*k',y) in the comprehensive feature map, r is the expansion rate, k' is the index of the convolution kernel used in the Atrous convolution, and k'=1,2,…,K, K represents the number of convolution kernels used in the Atrous convolution, W k' I2 represents the weight of the k'th convolution kernel in the Atrous convolution, I2 represents the second feature map, and I2(x,y) represents the pixel value at position (x,y) in the second feature map.

[0027] 1.4 Based on the channel attention mechanism in the CBAM attention module, the second feature map is processed to generate a channel attention map, using the following formula:

[0028]

[0029] Among them, F avg (I2) represents the second feature map after global average pooling operation, F max (I2) represents the second feature map after global max pooling, where σ is the sigmoid activation function, and W... c Let be the weight of the c-th channel in the channel attention mechanism, and M is the channel attention map obtained after processing by the channel attention mechanism;

[0030] 1.5 Based on the spatial attention mechanism in the CBAM attention module, the comprehensive feature map, channel attention map, and second feature map are processed to generate a spatial attention map. The formula used is as follows:

[0031] M s =σ(Conv 7×7 (Concat(M*I2,M*Iz)))

[0032] Here, `Concat(·)` indicates performing a convolution operation using a 7×7 convolution kernel, and `Concat(M*I2, M*Iz)` indicates concatenating `M*I2` and `M*Iz`. s The spatial attention map obtained after processing by the spatial attention mechanism;

[0033] 1.6. Combining the channel attention map, spatial attention map, and second feature map, a segmentation feature map is generated based on the following formula:

[0034] O=M s *(M*I2)

[0035] Where O represents the feature map used for segmentation;

[0036] 1.7 The segmentation feature map is fed into the classification layer of the image segmentation model for training to generate predicted segmentation results. The predicted segmentation results are compared with the labeled actual segmentation results. The loss function value is calculated based on the cross-entropy loss function. The undetermined values ​​used when generating the segmentation feature map are optimized through the backpropagation algorithm. The loss function value is minimized through multiple iterations of training until training is completed.

[0037] Furthermore, the formula for calculating the intersection-union ratio is as follows:

[0038]

[0039] Among them, A l A represents the area of ​​the impact point image. m A represents the area of ​​the target image that needs to be hit. j This represents the area of ​​intersection between the image of the impact point and the image of the target to be hit.

[0040] Furthermore, step S4 includes the following steps:

[0041] S41, the target tracking module based on the photoelectric gimbal obtains the three-dimensional coordinates of the target to be hit in real time, projects the three-dimensional coordinates onto the screen coordinate system to obtain the two-dimensional coordinates of the target to be hit, and determines the center coordinates of the impact point area of ​​the impact point image on the same screen coordinate system.

[0042] The projection formula used to project the three-dimensional coordinates of the target to two-dimensional coordinates is as follows:

[0043]

[0044] Among them, f x ,f y All are the focal lengths of the cameras on the optoelectronic gimbal, f x f is used to indicate the magnification of a camera in the horizontal direction. y c is used to indicate the magnification of a camera in the vertical direction. x ,c y Principal point coordinates, X m ,Y m Z m The target tracking module acquires the three-dimensional coordinates of the target to be attacked in real time, X m Y represents the horizontal coordinates of the target in the camera coordinate system. m Z represents the vertical coordinates of the target in the camera coordinate system. m The depth coordinates are the distance between the target and the camera. The depth coordinates represent the distance of the target in front of the camera. m ,y mLet x be a two-dimensional coordinate projected onto the screen coordinate system. m Let y be the horizontal coordinate of the target in the screen coordinate system. m This refers to the vertical coordinates of the target in the screen coordinate system.

[0045] S42, calculate the pixel distance deviation between the center coordinates of the impact area and the two-dimensional coordinates of the target to be hit. The calculation formula is as follows:

[0046]

[0047] Where, x l Let y be the horizontal coordinate of the center of the landing area in the screen coordinate system. l Δx represents the vertical coordinate of the center of the landing area in the screen coordinate system, Δy represents the pixel distance deviation in the horizontal direction, and Δy represents the pixel distance deviation in the vertical direction.

[0048] S43, calculates the object distance deviation based on pixel distance deviation, using the following formula:

[0049]

[0050] Where, d x d represents the horizontal distance deviation of the object. y This represents the deviation in actual distance in the vertical direction;

[0051] S44, combining the actual distance deviation and the water cannon distance deviation, generates a comprehensive distance deviation, calculated using the following formula:

[0052]

[0053] Among them, dl x dl represents the horizontal distance deviation of the water cannon. y For the vertical water cannon distance deviation, dz x The overall distance deviation in the horizontal direction, dz y This represents the overall distance deviation in the vertical direction.

[0054] Furthermore, step S5 includes the following steps:

[0055] S51 converts the overall distance deviation into angular deviation using the following formula:

[0056]

[0057] Where L is the actual distance from the water cannon platform to the target to be struck, and θ x θ represents the yaw angle deviation. y This refers to the pitch angle deviation.

[0058] S52 sets the maximum single adjustment angle for yaw and pitch angles. Iteratively adjusts the yaw and pitch angles of the water cannon platform by combining the angle deviation and the maximum single adjustment angle. After each adjustment, the cross-comparison ratio is recalculated until the cross-comparison ratio is greater than the cross-comparison ratio threshold. During a single adjustment, the absolute values ​​of the yaw angle deviation and the pitch angle deviation are compared. If the absolute value of the yaw angle deviation is greater than the absolute value of the pitch angle deviation, the yaw angle is adjusted first; otherwise, the pitch angle is adjusted first.

[0059] Furthermore, the specific logic for adjusting the yaw and pitch angles of the water cannon platform by combining the angle deviation and the maximum single adjustment angle is as follows:

[0060] When the absolute value of the yaw angle deviation is greater than the maximum single adjustment angle of the yaw angle, the yaw angle is adjusted according to the maximum single adjustment angle of the yaw angle. When the absolute value of the yaw angle deviation is less than the maximum single adjustment angle of the yaw angle, the yaw angle is adjusted according to the yaw angle deviation. When the absolute value of the pitch angle deviation is greater than the maximum single adjustment angle of the pitch angle, the pitch angle is adjusted according to the maximum single adjustment angle of the pitch angle. When the absolute value of the pitch angle deviation is less than the maximum single adjustment angle of the pitch angle, the pitch angle is adjusted according to the vertical angle deviation.

[0061] A shipborne watergun platform adjustment device based on jet deviation prediction and ship attitude prediction, used to execute the above-mentioned shipborne watergun platform adjustment method based on jet deviation prediction and ship attitude prediction, comprising:

[0062] The data acquisition module is used to record the water cannon's impact area in real time to obtain real-time video frame images of the water cannon's impact area. The real-time video frame images are processed to obtain the real-time brightness value, real-time wave height, and real-time wave frequency of the water cannon's impact area, as well as the real-time air humidity and real-time visibility at the water cannon platform.

[0063] The image segmentation module is used to establish an image segmentation model that incorporates brightness value, wave height, wave frequency, air humidity and visibility, and inputs the data collected in step S1 into the image segmentation model to segment the water column image, the impact point image and the target image to be hit.

[0064] The strike judgment module is used to calculate the cross-union ratio between the strike point image and the target image. If the cross-union ratio is greater than the cross-union ratio threshold, it continues to fire at the current muzzle angle. If the cross-union ratio is not greater than the cross-union ratio threshold, it proceeds to step S4.

[0065] The deviation calculation module is used to obtain the three-dimensional coordinates of the target to be hit based on target tracking technology, and process the three-dimensional coordinates of the target to be hit and the impact point image to generate the physical distance deviation. Based on ship attitude prediction technology, it predicts the water cannon distance deviation between the water cannon and the current water cannon at the next water cannon strike. Combining the physical distance deviation and the water cannon distance deviation, a comprehensive distance deviation is generated.

[0066] The water cannon platform adjustment module is used to iteratively adjust the yaw and pitch angles of the water cannon platform based on the comprehensive distance deviation until the crossover ratio is greater than the crossover ratio threshold.

[0067] Compared with the prior art, the beneficial effects of the present invention are:

[0068] The present invention relates to a method and apparatus for adjusting a shipborne water cannon platform based on jet deviation prediction and ship attitude prediction. During image segmentation, an image segmentation model incorporating brightness, wave height, wave frequency, air humidity, and visibility is used to segment the image, fully considering the influence of external environmental factors. This results in more accurate segmented images of the water column, the impact point, and the target, thereby improving the accuracy of subsequent deviation calculations and water cannon platform angle adjustment, and ensuring precise adjustment of the water cannon platform to strike the target. Attached Figure Description

[0069] Figure 1 This is a flowchart of the shipborne water cannon platform adjustment method based on jet deviation prediction and ship attitude prediction in this invention.

[0070] Figure 2 This is a block diagram of the shipborne water cannon platform adjustment device based on jet deviation prediction and ship attitude prediction in this invention. Detailed Implementation

[0071] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0072] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0073] Example 1:

[0074] Please see Figure 1 This invention provides a method for adjusting a shipborne water cannon platform based on jet deviation prediction and ship attitude prediction, comprising the following steps:

[0075] S1, record the water cannon strike area in real time to obtain real-time video frame images of the water cannon strike area, process the real-time video frame images to obtain the real-time brightness value, real-time wave height and real-time wave frequency of the water cannon strike area, and obtain the real-time air humidity and real-time visibility at the water cannon platform.

[0076] It should be noted that the acquisition of video frame images is an existing technology. Specifically, it involves using a high-definition, high-frame-rate video camera to record the area hit by the water cannon, and then extracting and saving the video frame images from the recorded video to obtain the video frame images. To ensure the quality of the video frame images, the resolution of the video camera must be at least 1080p and the frame rate must be at least 30 frames per second.

[0077] It should be noted that real-time air humidity can be obtained through a humidity sensor, and real-time visibility can be obtained through a single-wavelength visibility meter. These are existing technologies and will not be elaborated upon here.

[0078] It should be noted that the method for obtaining the real-time brightness value is as follows: the brightness values ​​of all pixels in the real-time video frame image are summed and the average value is calculated. This average value is the real-time brightness value of the water cannon's strike area. This is existing technology and will not be elaborated here.

[0079] It should be noted that the method for obtaining the real-time wave height is as follows: based on the Canny edge detection algorithm, the peaks and troughs in the real-time video frame image are identified, and then the height difference between the peaks and troughs is calculated. This height difference is the real-time wave height. This is existing technology and will not be elaborated here.

[0080] It should be noted that the method for obtaining the real-time wave frequency is as follows: based on the Canny edge detection algorithm, the wave peaks in all video frame images within a unit time before real-time acquisition are identified, and the number of times the wave peaks appear within a unit time is counted. The ratio of this number to the unit time is the real-time wave frequency. The value of the unit time can be 1 second, 2 seconds, 3 seconds, etc., and the specific setting is determined by the staff according to the actual situation. There are no restrictions here.

[0081] S2, establish an image segmentation model that incorporates brightness value, wave height, wave frequency, air humidity and visibility, and input the data collected in step S1 into the image segmentation model to segment the water column image, the impact point image and the target image to be hit, including the following steps;

[0082] S21. An image segmentation model is established based on the deeplabv3+ model. The lightweight network MobileNetV2 is selected as the backbone network, and the CBAM attention module and data preprocessing module are inserted into the image segmentation model.

[0083] S22, Obtain the historical sample set, which includes historical video frame images, historical brightness values, historical wave heights, historical wave frequencies, historical air temperatures, and historical visibility. The historical sample set is randomly divided into a training set and a test set. Water columns, impact points, and targets to be hit are marked on the historical video frame images in the training set.

[0084] It should be noted that the historical video frame images, historical brightness values, historical wave heights, historical wave frequencies, historical air temperatures, and historical visibility in the historical sample set are all obtained using the same method as in step S1 above, and will not be elaborated here.

[0085] S23, input the historical sample set into the preprocessing module of the image segmentation model for preprocessing. The preprocessing includes denoising the video frame image and normalizing the video frame image, brightness value, wave height, wave frequency, air temperature and visibility.

[0086] It should be noted that Gaussian blurring, a technique already in use, can be employed for denoising video frame images. By using Gaussian blurring, noise in the video frame images is reduced, making them smoother and more pleasing, facilitating subsequent segmentation. Normalization of the video frame images can utilize existing techniques such as maximum-minimum normalization and mean normalization, normalizing the pixel values ​​to the range of 0-1. This improves the convergence speed of the subsequent image segmentation model, thereby increasing the rate of segmentation and laying a favorable foundation for real-time adjustment of the water cannon platform. For brightness, wave height, wave frequency, air temperature, and visibility, maximum-minimum normalization is used to normalize the data to the range of 0-1, further improving the convergence speed of the subsequent image segmentation model. When normalizing real-time data, the maximum and minimum values ​​from historical data are selected.

[0087] S24, the image segmentation model is trained based on the preprocessed training set, and the trained image segmentation model is tested based on the preprocessed test set until a qualified image segmentation model is obtained. The real-time data collected in step S1 is input into the qualified image segmentation model to segment the water column image, the impact point image and the target image to be hit.

[0088] The process of training the image segmentation model using the preprocessed training set includes the following steps:

[0089] 1.1 Feature extraction of video frame images based on MobileNetV2 to generate the first feature map. Specifically, depthwise separable convolution technology can be used to extract features from video frame images, based on the following formula:

[0090]

[0091] It should be noted that I represents the preprocessed video frame image, x is the horizontal coordinate of the current pixel, y is the vertical coordinate of the current pixel, k is the radius of the depthwise convolution kernel, i represents the horizontal offset of the current pixel in the depthwise convolution kernel, and i ranges from -k to k, j represents the vertical offset of the current pixel in the depthwise convolution kernel, and j ranges from -k to k, c represents the index of the image channel, and c = 1, 2, ..., C, where C is the number of channels of the image, I(x+i, y+j, c) represents the pixel value of the preprocessed video frame image at position (x+i, y+j) and channel c. In this embodiment, the acquired video frame image is an RGB image, and the number of channels C of the image is 3. c(i,j) represents the weight value of the depthwise convolution kernel at position (i,j) on channel c. Its value is continuously optimized during training to improve model performance. W(c) represents the weight value of the pointwise convolution on the c-th channel, used for linear combination of the feature maps of each channel. Its value is also continuously optimized during training to improve model performance. c I1 is the feature map obtained after the preprocessed video frame image is processed by depth convolution on channel c, and I2 is the first feature map obtained after depth-separable convolution.

[0092] 1.2 Based on the preprocessed brightness value, wave height, wave frequency, air humidity, and visibility, an environmental impact value is generated. The calculation formula is as follows:

[0093]

[0094] It should be noted that E represents the environmental impact value, which comprehensively considers the influence of five environmental factors: brightness, wave height, wave frequency, air humidity, and visibility. By integrating brightness, wave height, wave frequency, air humidity, and visibility into the environmental impact value, the impact of environmental factors on image segmentation is taken into account, while reducing the computational load and improving the segmentation efficiency of the subsequent image segmentation model. L, V, RH, H, f represent the preprocessed brightness, visibility, air humidity, wave height, and wave frequency, respectively, and V0 and RH0 represent the visibility threshold, air humidity threshold, and visibility threshold, respectively. The specific air humidity threshold is set by the staff according to the actual situation. It can be set according to the distance to the target to be hit. For example, the visibility threshold can be set to 30m and the air humidity threshold can be set to 40%. There are no restrictions here. λ1, λ2, λ3, λ4, C1, γ, δ, ω1, ω2 are all undetermined coefficients, and ω1+ω2=1. The method for determining these undetermined coefficients is to adjust them through training feedback during the image segmentation model training process. Specifically, the backpropagation algorithm is used for step-by-step optimization. This is existing technology and will not be elaborated here.

[0095] It should be noted that brightness value is an important factor affecting the image quality of video frames, and is especially important for image segmentation. The logarithmic function can handle the non-linear characteristics of brightness value, especially when the brightness value is low, the change of the logarithmic function is more sensitive. Therefore, the logarithmic function ln(L+1) is used to fully reflect the influence of brightness value on the image quality of video frames. The undetermined coefficient λ1 is used to adjust the weight of brightness value in the environmental influence value.

[0096] It should be noted that visibility affects the sharpness of the target in a video frame. Higher visibility results in better segmentation of the video frame. The sigmoid function is then used. To handle visibility, we can better capture its impact on segmentation results. The undetermined coefficient γ is used to control the Sigmoid function. The slope of the sigmoid function determines the strength of the influence of visibility on the segmentation effect. The visibility threshold V0 defines the sigmoid function. The midpoint of the image segmentation quality is adjusted by adjusting the undetermined coefficient γ and the visibility threshold V0. The undetermined coefficient λ2 is used to adjust the weight of visibility in the environmental impact value.

[0097] It should be noted that air humidity affects the propagation of light, especially near water surfaces. High humidity can cause image blurring, which in turn negatively impacts subsequent image segmentation. Therefore, a power-law function (RH - RH0) is used. δ To reflect the impact of air humidity on image segmentation, the undetermined coefficient δ determines the degree of influence of air humidity on the segmentation effect, and the air humidity threshold RH0 emphasizes the degree of influence of air humidity on the segmentation effect within a specific range. For example, when the air humidity exceeds the air humidity threshold, the impact on the image segmentation effect will increase significantly. The degree of influence of air humidity on the image segmentation quality can be adjusted by adjusting the undetermined coefficient δ and the air humidity threshold RH0. The undetermined coefficient λ3 is used to adjust the weight of air humidity in the environmental influence value.

[0098] It should be noted that wave height and wave frequency are important components of water surface disturbance, which directly affects the refraction and reflection of light. High water surface disturbance can lead to image blurring and interfere with image segmentation. Therefore, ω1*H+ω2*f is used to quantitatively evaluate water surface disturbance, and an exponential function e(ω1*H+ω2*f) is used to reflect the impact of water surface disturbance on image segmentation. As the water surface disturbance increases, its impact on image segmentation will increase significantly. The undetermined coefficient λ4 is used to adjust the weight of air humidity in the environmental impact value.

[0099] It should be noted that the undetermined coefficient C1 is a correction constant used to make an overall correction to the environmental impact value;

[0100] 1.3. The first feature map and the environmental impact value are concatenated to form a comprehensive feature map, and the comprehensive feature map is then subjected to Atrous convolution to obtain the second feature map. The formula used is as follows:

[0101]

[0102] It should be noted that Concat(I1,E) represents concatenating the first feature map and the environmental impact value. Specifically, TensorFlow or PyTorch techniques can be used to concatenate the first feature map and the environmental impact value. When using TensorFlow, the tf.expand_dims technique is used to first expand the environmental impact value so that it is expanded to the same spatial dimension as the first feature map before concatenating the two. When using PyTorch, the unsqueeze technique is used to first expand the environmental impact value so that it is expanded to the same spatial dimension as the first feature map before concatenating the two. Iz represents the comprehensive feature map formed by concatenating the first feature map and the environmental impact value. By concatenating the first feature map and the environmental impact value, a comprehensive feature map that takes into account the environmental impact is formed. When performing image segmentation in the subsequent process, the environmental impact factors are taken into account, thereby improving the accuracy of subsequent image segmentation.

[0103] It should be noted that Iz(x+r*k',y) represents the pixel value at position (x+r*k',y) in the comprehensive feature map, r is the expansion rate (e.g., when r=2, it means skipping 1 pixel for convolution), k' is the index of the convolution kernel used in the Atrous convolution, and k'=1,2,…,K, where K represents the number of convolution kernels used in the Atrous convolution, W k' I represents the weight of the k'th convolutional kernel in the Atrous convolution, which is continuously optimized during training to improve model performance. I2 represents the second feature map, and I2(x,y) represents the pixel value at position (x,y) in the second feature map.

[0104] 1.4 Based on the channel attention mechanism in the CBAM attention module, the second feature map is processed to generate a channel attention map, using the following formula:

[0105]

[0106] It should be noted that F avg (I2) represents the second feature map after global average pooling operation, F max (I2) represents the second feature map after global max pooling. Both global average pooling and global max pooling can be performed on the second feature map using existing techniques, which will not be elaborated upon here. σ is the sigmoid activation function, and W... c This represents the weight of the c-th channel in the channel attention mechanism. Its value is continuously optimized during training to improve model performance. M is the channel attention map obtained after processing by the channel attention mechanism;

[0107] 1.5 Based on the spatial attention mechanism in the CBAM attention module, the comprehensive feature map, channel attention map, and second feature map are processed to generate a spatial attention map. The formula used is as follows:

[0108] M s =σ(Conv 7×7 (Concat(M*I2,M*Iz)))

[0109] It should be noted that Concat(·) indicates that a 7×7 convolution kernel is used for the convolution operation, M s The spatial attention map obtained after processing by the spatial attention mechanism is first weighted by the second feature map and the comprehensive feature map through the channel attention map to form two weighted feature maps. Then, the two weighted feature maps are concatenated. Finally, the spatial attention mechanism is used to perform a convolution operation on the concatenated weighted feature map so that the final spatial attention map takes into account both channel and environmental factors.

[0110] 1.6. Combining the channel attention map, spatial attention map, and second feature map, a segmentation feature map is generated based on the following formula:

[0111] O=M s *(M*I2)

[0112] It should be noted that O represents the segmentation feature map. The second feature map is adjusted by the channel attention map and the spatial attention map to obtain the final segmentation feature map. The image segmentation model then analyzes and identifies different regions based on the segmentation feature map to improve the convergence speed of the image segmentation model during training.

[0113] 1.7 The segmentation feature map is fed into the classification layer of the image segmentation model for training to generate predicted segmentation results. The predicted segmentation results are compared with the labeled actual segmentation results. The loss function value is calculated based on the cross-entropy loss function. The undetermined values ​​used when generating the segmentation feature map are optimized through the backpropagation algorithm. The loss function value is minimized through multiple iterations of training until training is completed.

[0114] Among them, the undetermined values ​​include, but are not limited to, W. c (i,j),W(c),V0,RH0,λ1,λ2,λ3,λ4,C1,γ,δ,ω1,ω2,W k' 、W c ;

[0115] It should be noted that existing technologies can be used to train the image segmentation model using features extracted from the original data (such as the segmentation feature map in this embodiment). For example, the classification layer of the image segmentation model can be a fully connected layer, the loss function value calculated based on the cross-entropy loss function is an existing technology, and the parameter optimization based on the backpropagation algorithm is also an existing technology. The termination condition for training can be reaching a predetermined number of iterations or the loss function value being less than a preset threshold. The specific settings can be made by the staff according to the actual situation, and no restrictions are imposed here.

[0116] The standard for passing the image segmentation model test is that the accuracy of the water column image, impact point image, and target image segmented by the image segmentation model is higher than 92%, 95%, or 97%, etc. The specific settings are determined by the staff according to the actual situation and are not restricted here. If the test fails, the training steps are returned for retraining until the test is passed.

[0117] S3, calculate the cross-union ratio between the impact point image and the target image. If the cross-union ratio is greater than the cross-union ratio threshold, continue firing at the current muzzle angle. If the cross-union ratio is not greater than the cross-union ratio threshold, proceed to step S4.

[0118] The formula for calculating the intersection-union ratio is as follows:

[0119]

[0120] It should be noted that A l A represents the area of ​​the impact point image. m A represents the area of ​​the target image that needs to be hit. j A represents the area of ​​intersection between the image of the impact point and the image of the target to be hit. l A m A j All of these can be obtained through image processing software, which will not be elaborated here. IoU represents the intersection-to-union ratio, and the larger the intersection-to-union ratio IoU, the higher the overlap between the impact point and the target to be hit, which means that the water cannon is more accurate.

[0121] As one implementation method, the cross-union ratio threshold is set to 0.7. When the cross-union ratio IoU = 0, it means that the water jet did not hit the target. When the cross-union ratio IoU is greater than 0.7, it is considered that the target was hit completely, and the firing continues at the current muzzle angle. When the cross-union ratio IoU is less than 0.7, it is considered that the target was not hit completely or was not hit, and then step S4 is entered to calculate the actual distance deviation. Of course, the cross-union ratio threshold can also be set to other values ​​between 0 and 1. The specific setting is determined by the staff according to the actual situation, and no restrictions are imposed here.

[0122] S4. Based on target tracking technology, the three-dimensional coordinates of the target to be struck are obtained. The three-dimensional coordinates of the target to be struck and the impact point image are processed to generate the actual distance deviation. Based on ship attitude prediction technology, the water cannon distance deviation between the current water cannon and the next water cannon strike is predicted. Combining the actual distance deviation and the water cannon distance deviation, a comprehensive distance deviation is generated, including the following steps:

[0123] S41, the target tracking module based on the photoelectric gimbal obtains the three-dimensional coordinates of the target to be hit in real time, projects the three-dimensional coordinates onto the screen coordinate system to obtain the two-dimensional coordinates of the target to be hit, and determines the center coordinates of the impact point area of ​​the impact point image on the same screen coordinate system.

[0124] It should be noted that determining the center coordinates of the impact area is existing technology and will not be elaborated here. The optoelectronic gimbal can use a shipborne dual-spectrum thermal imaging night vision gimbal camera. The target tracking module is a built-in module of the optoelectronic gimbal. The real-time acquisition of the three-dimensional coordinates of the target to be hit based on the target tracking module of the optoelectronic gimbal is existing technology and will not be elaborated here.

[0125] Among them, projecting the three-dimensional coordinates of the target to be struck into two-dimensional coordinates is a current technology, and the projection formula used is as follows:

[0126]

[0127] It should be noted that f x ,f y All are the focal lengths of the camera on the optoelectronic gimbal, where f x f is used to indicate the magnification of a camera in the horizontal direction. y f is used to indicate the magnification of a camera in the vertical direction. x ,f y All of these can be obtained from the parameter table of the camera on the opto-panel head. x ,c y The principal point coordinates are specifically the coordinates of the projection of the camera coordinate system origin onto the screen coordinate system, X. m ,Y m Z m The target tracking module acquires the three-dimensional coordinates of the target to be attacked in real time, where X... m Y represents the horizontal coordinates of the target in the camera coordinate system. m Z represents the vertical coordinates of the target in the camera coordinate system. m The depth coordinates are the distance between the target and the camera. The depth coordinates represent the distance of the target in front of the camera. m ,y m Let x be a two-dimensional coordinate projected onto the screen coordinate system, where x m Let y be the horizontal coordinate of the target in the screen coordinate system.m This refers to the vertical coordinates of the target in the screen coordinate system.

[0128] S42, calculate the pixel distance deviation between the center coordinates of the impact area and the two-dimensional coordinates of the target to be hit. The calculation formula is as follows:

[0129]

[0130] It should be noted that x l Let y be the horizontal coordinate of the center of the landing area in the screen coordinate system. l Δx is the vertical coordinate of the center of the landing area in the screen coordinate system, Δy is the pixel distance deviation in the horizontal direction, which is used to represent the number of pixels that need to be adjusted in the horizontal direction, and Δy is the pixel distance deviation in the vertical direction, which is used to represent the number of pixels that need to be adjusted in the vertical direction.

[0131] S43, calculates the object distance deviation based on pixel distance deviation, using the following formula:

[0132]

[0133] It should be noted that d x d represents the deviation of the actual distance in the horizontal direction, used to indicate the actual distance that needs to be adjusted in the horizontal direction. y This represents the deviation of the actual distance in the vertical direction, used to indicate the actual distance that needs to be adjusted in the vertical direction.

[0134] S44, combining the actual distance deviation and the water cannon distance deviation, generates a comprehensive distance deviation, calculated using the following formula:

[0135]

[0136] It should be noted that dl x dl represents the horizontal distance deviation of the water cannon. y The vertical water cannon distance deviation, and the horizontal and vertical water cannon distance deviation dl x and dl y All data are obtained through ship attitude prediction technology. Specifically, according to the patent technology disclosed in CN115950395A, a method for ship attitude prediction and shipborne water cannon muzzle angle adjustment, the coordinates of the water cannon platform at the time of the next water cannon strike and the current coordinates of the water cannon platform are obtained. Subtracting the current coordinates from the coordinates of the water cannon platform at the time of the next water cannon strike yields the horizontal and vertical water cannon distance deviations dl. x And dly, this is existing technology, which will not be elaborated here, and the transformation of three-dimensional coordinates and two-dimensional coordinates used in this process can adopt the technology in this embodiment or existing technology, and there is no limitation here. dzx The overall distance deviation in the horizontal direction, dz y This represents the overall distance deviation in the vertical direction.

[0137] S5, based on the comprehensive distance deviation, iteratively adjust the yaw and pitch angles of the water cannon platform until the crossover ratio is greater than the crossover ratio threshold, including the following steps:

[0138] S51 converts the overall distance deviation into angular deviation using the following formula:

[0139]

[0140] It should be noted that L represents the actual distance from the water cannon platform to the target, which can be obtained using ranging equipment such as a laser rangefinder. θ x θ represents the yaw angle deviation. y This refers to the pitch angle deviation.

[0141] S52 sets the maximum single adjustment angle for yaw and pitch angles. Iteratively adjusts the yaw and pitch angles of the water cannon platform by combining the angle deviation and the maximum single adjustment angle. After each adjustment, the cross-comparison ratio is recalculated until the cross-comparison ratio is greater than the cross-comparison ratio threshold. During a single adjustment, the absolute values ​​of the yaw angle deviation and the pitch angle deviation are compared. If the absolute value of the yaw angle deviation is greater than the absolute value of the pitch angle deviation, the yaw angle is adjusted first; otherwise, the pitch angle is adjusted first.

[0142] The method for adjusting the yaw and pitch angles of the water cannon platform by combining angle deviation and single maximum adjustment angle is as follows: when the absolute value of the yaw angle deviation is greater than the single maximum adjustment angle of the yaw angle, the yaw angle is adjusted according to the single maximum adjustment angle of the yaw angle; when the absolute value of the yaw angle deviation is less than the single maximum adjustment angle of the yaw angle, the yaw angle is adjusted according to the yaw angle deviation. When the absolute value of the pitch angle deviation is greater than the single maximum adjustment angle of the pitch angle, the pitch angle is adjusted according to the single maximum adjustment angle of the pitch angle; when the absolute value of the pitch angle deviation is less than the single maximum adjustment angle of the pitch angle, the pitch angle is adjusted according to the vertical angle deviation. Since there is a certain error in the conversion between two-dimensional coordinates and three-dimensional coordinates, a large single rotation angle will lead to inaccurate results. Here, an iterative adjustment algorithm is used to effectively improve the adjustment accuracy.

[0143] It should be noted that the maximum single adjustment angle for both yaw and pitch can be 3, 4, 5, or 7 degrees, etc., and the specific setting is determined by the staff based on the actual situation. No restrictions are imposed here.

[0144] Example 2:

[0145] Please see Figure 2This invention provides a shipborne water cannon platform adjustment device based on jet deviation prediction and ship attitude prediction, used to execute the shipborne water cannon platform adjustment method based on jet deviation prediction and ship attitude prediction in Embodiment 1 above, including:

[0146] The data acquisition module is used to record the water cannon's impact area in real time to obtain real-time video frame images of the water cannon's impact area. The real-time video frame images are processed to obtain the real-time brightness value, real-time wave height, and real-time wave frequency of the water cannon's impact area, as well as the real-time air humidity and real-time visibility at the water cannon platform.

[0147] The image segmentation module is used to establish an image segmentation model that incorporates brightness value, wave height, wave frequency, air humidity and visibility, and inputs the data collected in step S1 into the image segmentation model to segment the water column image, the impact point image and the target image to be hit.

[0148] The strike judgment module is used to calculate the cross-union ratio between the strike point image and the target image. If the cross-union ratio is greater than the cross-union ratio threshold, it continues to fire at the current muzzle angle. If the cross-union ratio is not greater than the cross-union ratio threshold, it proceeds to step S4.

[0149] The deviation calculation module is used to obtain the three-dimensional coordinates of the target to be hit based on target tracking technology, and process the three-dimensional coordinates of the target to be hit and the impact point image to generate the physical distance deviation. Based on ship attitude prediction technology, it predicts the water cannon distance deviation between the water cannon and the current water cannon at the next water cannon strike. Combining the physical distance deviation and the water cannon distance deviation, a comprehensive distance deviation is generated.

[0150] The water cannon platform adjustment module is used to iteratively adjust the yaw and pitch angles of the water cannon platform based on the comprehensive distance deviation until the crossover ratio is greater than the crossover ratio threshold.

[0151] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0152] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0153] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0154] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. A method for adjusting a shipborne water cannon platform based on jet deviation prediction and ship attitude prediction, characterized in that, The specific steps include: S1, record the water cannon strike area in real time to obtain real-time video frame images of the water cannon strike area, process the real-time video frame images to obtain the real-time brightness value, real-time wave height and real-time wave frequency of the water cannon strike area, and obtain the real-time air humidity and real-time visibility at the water cannon platform. S2, establish an image segmentation model that incorporates brightness value, wave height, wave frequency, air humidity and visibility, and input the data collected in step S1 into the image segmentation model to segment the water column image, the impact point image and the target image to be hit; S3, calculate the cross-union ratio between the impact point image and the target image. If the cross-union ratio is greater than the cross-union ratio threshold, continue firing at the current muzzle angle. If the cross-union ratio is not greater than the cross-union ratio threshold, proceed to step S4. S4. Based on target tracking technology, the three-dimensional coordinates of the target to be hit are obtained, and the three-dimensional coordinates of the target to be hit and the impact point image are processed to generate the physical distance deviation. Based on ship attitude prediction technology, the water cannon distance deviation between the water cannon and the current water cannon at the next water cannon strike is predicted. The physical distance deviation and the water cannon distance deviation are combined to generate the comprehensive distance deviation. S5, based on the comprehensive distance deviation, iteratively adjusts the yaw and pitch angles of the water cannon platform until the crossover ratio is greater than the crossover ratio threshold.

2. The method for adjusting a shipborne water cannon platform based on jet deviation prediction and ship attitude prediction according to claim 1, characterized in that: Step S2 includes the following steps; S21. An image segmentation model is established based on the deeplabv3+ model. The lightweight network MobileNetV2 is selected as the backbone network, and the CBAM attention module and data preprocessing module are inserted into the image segmentation model. S22, Obtain the historical sample set, which includes historical video frame images, historical brightness values, historical wave heights, historical wave frequencies, historical air temperatures, and historical visibility. The historical sample set is randomly divided into a training set and a test set. Water columns, impact points, and targets to be hit are marked on the historical video frame images in the training set. S23, input the historical sample set into the preprocessing module of the image segmentation model for preprocessing. The preprocessing includes denoising the video frame image and normalizing the video frame image, brightness value, wave height, wave frequency, air temperature and visibility. S24, the image segmentation model is trained based on the preprocessed training set, and the trained image segmentation model is tested based on the preprocessed test set until a qualified image segmentation model is obtained. The real-time data collected in step S1 is input into the qualified image segmentation model to segment the water column image, the impact point image, and the target image to be hit.

3. The method for adjusting a shipborne water cannon platform based on jet deviation prediction and ship attitude prediction according to claim 2, characterized in that: The specific logic for training the image segmentation model using the preprocessed training set is as follows: 1.1, Feature extraction is performed on video frame images based on depthwise separable convolution technology to generate a first feature map; 1.2 Based on the preprocessed brightness value, wave height, wave frequency, air humidity, and visibility, an environmental impact value is generated. The calculation formula is as follows: Where E is the environmental impact value, L, V, RH, H, f are the pre-processed brightness value, visibility, air humidity, wave height and wave frequency respectively, V0 and RH0 are the visibility threshold and air humidity threshold respectively, λ1, λ2, λ3, λ4, C1, γ, δ, ω1, ω2 are all undetermined coefficients, and ω1+ω2=1. 1.

3. The first feature map and the environmental impact value are concatenated to form a comprehensive feature map, and the comprehensive feature map is then subjected to Atrous convolution to obtain the second feature map. The formula used is as follows: Where I1 represents the first feature map, Concat(I1,E) concatenates the first feature map and the environmental influence value, Iz represents the comprehensive feature map, Iz(x+r*k',y) represents the pixel value at position (x+r*k',y) in the comprehensive feature map, r is the expansion rate, k' is the index of the convolution kernel used in the Atrous convolution, and k'=1,2,…,K, K represents the number of convolution kernels used in the Atrous convolution, W k' I2 represents the weight of the k'th convolution kernel in the Atrous convolution, I2 represents the second feature map, and I2(x,y) represents the pixel value at position (x,y) in the second feature map. 1.4 Based on the channel attention mechanism in the CBAM attention module, the second feature map is processed to generate a channel attention map, using the following formula: Among them, F avg (I2) represents the second feature map after global average pooling operation, F max (I2) represents the second feature map after global max pooling, where σ is the sigmoid activation function, and W... c Let be the weight of the c-th channel in the channel attention mechanism, and M is the channel attention map obtained after processing by the channel attention mechanism; 1.5 Based on the spatial attention mechanism in the CBAM attention module, the comprehensive feature map, channel attention map, and second feature map are processed to generate a spatial attention map. The formula used is as follows: M s =σ(Conv 7×7 (Concat(M*I2,M*Iz))) Here, `Concat(·)` indicates performing a convolution operation using a 7×7 convolution kernel, and `Concat(M*I2, M*Iz)` indicates concatenating `M*I2` and `M*Iz`. s The spatial attention map obtained after processing by the spatial attention mechanism; 1.

6. Combining the channel attention map, spatial attention map, and second feature map, a segmentation feature map is generated based on the following formula: O=M s *(M*I2) Where O represents the feature map used for segmentation; 1.7 The segmentation feature map is fed into the classification layer of the image segmentation model for training to generate predicted segmentation results. The predicted segmentation results are compared with the labeled actual segmentation results. The loss function value is calculated based on the cross-entropy loss function. The undetermined values ​​used when generating the segmentation feature map are optimized through the backpropagation algorithm. The loss function value is minimized through multiple iterations of training until training is completed.

4. The method for adjusting a shipborne water cannon platform based on jet deviation prediction and ship attitude prediction according to claim 1, characterized in that: The formula for calculating the crossover-union ratio is as follows: Among them, A l A represents the area of ​​the impact point image. m A represents the area of ​​the target image that needs to be hit. j This represents the area of ​​intersection between the image of the impact point and the image of the target to be hit.

5. The method for adjusting a shipborne water cannon platform based on jet deviation prediction and ship attitude prediction according to claim 1, characterized in that: Step S4 includes the following steps: S41, the target tracking module based on the photoelectric gimbal obtains the three-dimensional coordinates of the target to be hit in real time, projects the three-dimensional coordinates onto the screen coordinate system to obtain the two-dimensional coordinates of the target to be hit, and determines the center coordinates of the impact point area of ​​the impact point image on the same screen coordinate system. The projection formula used to project the three-dimensional coordinates of the target to two-dimensional coordinates is as follows: Among them, f x ,f y All are the focal lengths of the cameras on the optoelectronic gimbal, f x f is used to indicate the magnification of a camera in the horizontal direction. y c is used to indicate the magnification of a camera in the vertical direction. x ,c y Principal point coordinates, X m ,Y m Z m The target tracking module acquires the three-dimensional coordinates of the target to be attacked in real time, X m Y represents the horizontal coordinates of the target in the camera coordinate system. m Z represents the vertical coordinates of the target in the camera coordinate system. m The depth coordinates are the distance between the target and the camera. The depth coordinates represent the distance of the target in front of the camera. m ,y m Let x be a two-dimensional coordinate projected onto the screen coordinate system. m Let y be the horizontal coordinate of the target in the screen coordinate system. m This refers to the vertical coordinates of the target in the screen coordinate system. S42, calculate the pixel distance deviation between the center coordinates of the impact area and the two-dimensional coordinates of the target to be hit. The calculation formula is as follows: Where, x l y is the horizontal coordinate of the center of the landing area in the screen coordinate system, yl is the vertical coordinate of the center of the landing area in the screen coordinate system, Δx is the pixel distance deviation in the horizontal direction, and Δy is the pixel distance deviation in the vertical direction. S43, calculates the object distance deviation based on pixel distance deviation, using the following formula: Where, d x d represents the horizontal distance deviation of the object. y This represents the deviation in actual distance in the vertical direction; S44, combining the actual distance deviation and the water cannon distance deviation, generates a comprehensive distance deviation, calculated using the following formula: Among them, dl x dl represents the horizontal distance deviation of the water cannon. y For the vertical water cannon distance deviation, dz x The overall distance deviation in the horizontal direction, dz y This represents the overall distance deviation in the vertical direction.

6. The method for adjusting a shipborne water cannon platform based on jet deviation prediction and ship attitude prediction according to claim 1, characterized in that: Step S5 includes the following steps: S51 converts the overall distance deviation into angular deviation using the following formula: Where L is the actual distance from the water cannon platform to the target to be struck, and θ x θ represents the yaw angle deviation. y This refers to the pitch angle deviation. S52 sets the maximum single adjustment angle for yaw and pitch angles. Iteratively adjusts the yaw and pitch angles of the water cannon platform by combining the angle deviation and the maximum single adjustment angle. After each adjustment, the cross-comparison ratio is recalculated until the cross-comparison ratio is greater than the cross-comparison ratio threshold. During a single adjustment, the absolute values ​​of the yaw angle deviation and the pitch angle deviation are compared. If the absolute value of the yaw angle deviation is greater than the absolute value of the pitch angle deviation, the yaw angle is adjusted first; otherwise, the pitch angle is adjusted first.

7. The method for adjusting a shipborne water cannon platform based on jet deviation prediction and ship attitude prediction according to claim 6, characterized in that: The specific logic for adjusting the yaw and pitch angles of the water cannon platform by combining the angle deviation and the maximum adjustment angle in a single operation is as follows: When the absolute value of the yaw angle deviation is greater than the maximum single adjustment angle of the yaw angle, the yaw angle is adjusted according to the maximum single adjustment angle of the yaw angle. When the absolute value of the yaw angle deviation is less than the maximum single adjustment angle of the yaw angle, the yaw angle is adjusted according to the yaw angle deviation. When the absolute value of the pitch angle deviation is greater than the maximum single adjustment angle of the pitch angle, the pitch angle is adjusted according to the maximum single adjustment angle of the pitch angle. When the absolute value of the pitch angle deviation is less than the maximum single adjustment angle of the pitch angle, the pitch angle is adjusted according to the vertical angle deviation.

8. A shipborne water cannon platform adjustment device based on jet deviation prediction and ship attitude prediction, used to execute the shipborne water cannon platform adjustment method based on jet deviation prediction and ship attitude prediction as described in any one of claims 1-7, characterized in that, include: The data acquisition module is used to record the water cannon's impact area in real time to obtain real-time video frame images of the water cannon's impact area. The real-time video frame images are processed to obtain the real-time brightness value, real-time wave height, and real-time wave frequency of the water cannon's impact area, as well as the real-time air humidity and real-time visibility at the water cannon platform. The image segmentation module is used to establish an image segmentation model that incorporates brightness value, wave height, wave frequency, air humidity and visibility, and inputs the data collected in step S1 into the image segmentation model to segment the water column image, the impact point image and the target image to be hit. The strike judgment module is used to calculate the cross-union ratio between the strike point image and the target image. If the cross-union ratio is greater than the cross-union ratio threshold, it continues to fire at the current muzzle angle. If the cross-union ratio is not greater than the cross-union ratio threshold, it proceeds to step S4. The deviation calculation module is used to obtain the three-dimensional coordinates of the target to be hit based on target tracking technology, and process the three-dimensional coordinates of the target to be hit and the impact point image to generate the physical distance deviation. Based on ship attitude prediction technology, it predicts the water cannon distance deviation between the water cannon and the current water cannon at the next water cannon strike. Combining the physical distance deviation and the water cannon distance deviation, a comprehensive distance deviation is generated. The water cannon platform adjustment module is used to iteratively adjust the yaw and pitch angles of the water cannon platform based on the comprehensive distance deviation until the crossover ratio is greater than the crossover ratio threshold.

Images