Marine ship length measuring method of intelligent reconnaissance equipment

Abstract

The invention relates to the technical field of ship length measurement, and provides a maritime ship length measurement method of intelligent reconnaissance equipment, which comprises the following steps of: acquiring ship images, distance direction data and sensor attitude data of a maritime target ship through an image acquisition sensor and a laser distance measuring sensor on photoelectric pod equipment; inputting the ship image into a trained target detection model to obtain a ship identification frame; generating a ship three-dimensional point cloud according to the distance direction data; fusing the ship image, the ship three-dimensional point cloud and the ship identification frame based on a spatial transformation matrix to obtain a fused image; and inputting the fused image and the sensor attitude data into a trained deep neural network to obtain a ship length measurement result. According to the method, the accuracy of ship length measurement can be improved, meanwhile, the robustness and the real-time performance of the model are enhanced, and effective technical support is provided for monitoring and management of ships on the sea.

Description

Technical Field

[0001] The present invention relates to the technical field of ship length measurement, and more specifically, to a method for measuring the length of a marine ship using intelligent reconnaissance equipment. Background Art

[0002] The optoelectronic pod is an optoelectronic detection device widely used in military reconnaissance, terrain detection, patrol and security. At present, the research on the length and depth measurement of ships at sea by the traditional optoelectronic pod system mainly focuses on target recognition and coordinate transformation of multi-cameras, but lacks research on high-level semantic understanding of images and video streams, making it difficult to achieve in-depth analysis of targets and scenes. These limitations hinder the development of optoelectronic pod systems in dealing with complex dynamic environments and realizing intelligent processing.

[0003] With the emergence of deep learning, Convolutional Neural Networks (CNN) have made significant breakthroughs in automatic image interpretation in the field of maritime monitoring. However, in optoelectronic pod systems, the use of deep learning technology to measure the length of ships faces many challenges. First, since ships are relatively small in images, the available target features are limited, especially when shooting at long distances. In addition, the diversity of ships in the marine environment, changes in different perspectives and scales, and complex background conditions all increase the difficulty of detection. At the same time, there are currently proposed methods based on three-dimensional point cloud reconstruction and database matching for accurate measurement. However, in practical applications, these methods cannot meet the real-time requirements of ship length measurement due to their high computational cost and low efficiency, especially in scenarios where rapid measurement is required, their limitations are particularly obvious. Summary of the invention

[0004] In order to overcome the defects of low efficiency and low accuracy in the above-mentioned existing real-time ship length measurement and detection technology, the present invention provides a method for measuring the length of a marine ship using intelligent reconnaissance equipment.

[0005] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0006] A method for measuring the length of a marine vessel using intelligent reconnaissance equipment comprises the following steps:

[0007] The image acquisition sensor and laser ranging sensor on the optoelectronic pod device collect the ship image, distance direction data and sensor attitude data of the target ship at sea;

[0008] Inputting the ship image into a trained object detection model to obtain a ship recognition frame;

[0009] generating a three-dimensional point cloud of the vessel according to the distance direction data;

[0010] The ship image, the ship three-dimensional point cloud and the ship identification frame are fused based on a spatial transformation matrix to obtain a fused image;

[0011] The fused image and the sensor posture data are input into a trained deep neural network to obtain a ship length measurement result.

[0012] Furthermore, the present invention also proposes a marine vessel length measurement system of intelligent reconnaissance equipment, which applies the marine vessel length measurement method proposed in the present invention. The system comprises:

[0013] Intelligent reconnaissance equipment, including optoelectronic pod equipment and laser rangefinder sensors, used to collect ship images, distance direction data and sensor attitude data of target ships at sea;

[0014] An object detection module, which is equipped with a trained object detection model and is used to obtain a ship recognition frame based on an input ship image;

[0015] A point cloud generation module, used for generating a three-dimensional point cloud of a ship according to the distance direction data;

[0016] A fusion module, used for fusing the ship image, the ship three-dimensional point cloud and the ship identification frame based on a spatial transformation matrix to obtain a fused image;

[0017] A measurement module is equipped with a trained deep neural network and is used to output a ship length measurement result based on the input fused image and the sensor posture data.

[0018] Furthermore, the present invention also proposes a device, including a memory and a processor, wherein the memory stores computer-readable instructions, wherein when the computer-readable instructions are executed by the processor, the processor executes the steps of the anti-interference vehicle detection method proposed in the present invention.

[0019] Furthermore, the present invention also proposes a storage medium on which computer-readable instructions are stored, wherein the computer-readable instructions, when executed by a processor, implement the steps of the anti-interference vehicle detection method proposed in the present invention.

[0020] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0021] The present invention performs target detection on ship images to identify ship targets, combines the fusion data of three-dimensional point clouds and ship images, and uses sensor attitude data to assist in the estimation and measurement of ship length, so as to improve the accuracy of ship length measurement, enhance the robustness and real-time performance of the model, and provide effective technical support for marine ship monitoring and management. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 The figure is a flow chart of a method for measuring the length of a marine vessel according to an embodiment of the present invention.

[0023] Figure 2 The figure is a flow chart of a method for measuring the length of a marine vessel according to an embodiment of the present invention.

[0024] Figure 3 The figure is an architecture diagram of a mixed receptive field convolution according to an embodiment of the present invention.

[0025] Figure 4 The diagram is an architecture diagram of a coordinate attention module according to one embodiment of the present invention.

[0026] Figure 5 The figure is a schematic diagram of an R-Tree data structure according to an embodiment of the present invention.

[0027] Figure 6 FIG. 4 is a schematic diagram showing the result of measuring the length of a ship at sea according to an embodiment of the present invention.

[0028] Figure 7 FIG. 4 is a schematic diagram of a marine vessel length measurement system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0029] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0030] The terms used in the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The singular forms "a", "the" and "the" used in the present invention and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0031] It should be understood that although the terms first, second, third, etc. may be used in the present invention to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present invention, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0032] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0033] Example 1

[0034] This embodiment provides a method for measuring the length of a ship at sea using intelligent reconnaissance equipment. Figure 1 , 2 FIG. 1 is a flow chart of the method for measuring the length of a marine vessel according to the present embodiment.

[0035] The method for measuring the length of a ship at sea using intelligent reconnaissance equipment proposed in this embodiment includes the following steps:

[0036] S1. Collect ship images, distance direction data and sensor attitude data of target ships at sea through image acquisition sensors and laser ranging sensors on optoelectronic pod equipment;

[0037] S2, inputting the ship image into a trained target detection model to obtain a ship recognition frame;

[0038] S3, generating a three-dimensional point cloud of the ship according to the distance direction data;

[0039] S4, fusing the ship image, the ship three-dimensional point cloud and the ship identification frame based on a spatial transformation matrix to obtain a fused image;

[0040] S5. Input the fused image and the sensor posture data into a trained deep neural network to obtain a ship length measurement result.

[0041] In this embodiment, the ship target is identified by performing target detection on the ship image, combining the fusion data of the three-dimensional point cloud and the ship image, and using the sensor attitude data to assist in the estimation and measurement of the ship length, so as to improve the accuracy of the ship length measurement, and at the same time enhance the robustness and real-time performance of the model, thereby providing effective technical support for marine ship monitoring and management.

[0042] Exemplarily, this embodiment uses an optoelectronic pod device as the intelligent reconnaissance equipment, on which an image acquisition sensor and a laser ranging sensor are integrated, which are used to collect the operating status of the optoelectronic pod device, the optoelectronic pod angle data, the laser radar image and visible light image collected by the optoelectronic pod camera, etc. All or part of the steps of the method described in this embodiment are deployed in the processor or smart chip in the intelligent reconnaissance equipment, or a communication module is deployed in the intelligent reconnaissance equipment to transmit the collected data to the terminal, so as to execute all or part of the steps of the method.

[0043] Exemplarily, the optoelectronic pod equipment also includes a stabilizing platform, an inertial sensor and other equipment to ensure the stability of the equipment during data collection.

[0044] Exemplarily, the image acquisition sensor on the optoelectronic pod device may optionally use a MV-CH650-90XM camera to capture visible light images. The camera has a number of advanced features such as in-camera pattern correction, multiple ROIs, correlated double sampling, combined analog and digital gain control, etc., which can ensure effective noise reduction and provide high-quality images with high pixel uniformity.

[0045] Exemplarily, the laser ranging sensor on the optoelectronic pod device can optionally use an LF laser ranging module. Its features include: high accuracy, up to 1 mm; wide range, from 0.05 meters to 80 meters; fast frequency, up to 20Hz; suitable for indoor industrial measurement. The laser ranging module uses the flight time difference of the laser to measure the distance, and adopts the phase laser ranging method to convert the distance by measuring the phase delay between the laser emission and the echo signal. It can also be combined with a surveying camera to generate true color point cloud data.

[0046] In this embodiment, the image acquisition sensor and the laser ranging sensor collect data simultaneously and store the data in the memory.

[0047] In an optional embodiment, the target detection model includes a YOLOv5 network using mixed receptive field convolution.

[0048] In this embodiment, the expression of the t-th group output of the mixed receptive field convolution is:

[0049]

[0050] In the formula, represents the feature map of the t-th group output, represents the feature map of the tth group of input; h, w, c represent the height, width and channel of the input feature map respectively; represents the convolution kernel of the tth group; k t represents the convolution kernel size of the tth group; m is the multiplication factor of the channel, c tIndicates the number of channels.

[0051] Exemplarily, the convolution operation of the backbone network in the target detection model is based on the idea of ​​mixed receptive field convolution, which improves the original convolution kernel into a combination of 1×1 and 3×3 convolution kernels. Figure 3 As shown, it is the architecture diagram of the mixed receptive field convolution of this embodiment. Among them, the channels of the input feature map will be divided into several groups, which are calculated with several convolution kernels of different sizes respectively, and then the obtained output feature maps are concat-operated to obtain the final output feature map.

[0052] In this embodiment, by combining mixed receptive field convolution to build a target detection model, the advantages of multiple convolution kernels are fully utilized to better extract features. As an improvement on the ordinary convolution layer, it will improve the accuracy and efficiency of the target detection method. Since the convolution operation using multiple convolution kernels leads to inconsistent outputs, the target detection model adjusts them to the same size by filling zeros at the boundaries, and finally merges them through the Concat operation, thereby improving the performance of the model and the accuracy of target detection.

[0053] Furthermore, in an optional embodiment, the target detection model also includes a coordinate attention module; the coordinate attention module includes a residual network layer, an average pooling layer, a compression layer, a spatial information encoding layer, a convolution layer and a normalization layer connected in sequence.

[0054] Among them, after the input feature map is processed by the residual network layer, it enters the average pooling layer for average pooling in the horizontal and vertical directions respectively, to obtain a first one-dimensional vector and a second one-dimensional vector; the first one-dimensional vector and the second one-dimensional vector enter the compression layer for Concat and Conv2d operations to compress the number of channels, and then the spatial information in the vertical and horizontal directions is encoded by batch normalization and non-linear methods through the spatial information encoding layer, and is divided into a first tensor and a second tensor in the horizontal and vertical directions in the spatial dimension; the first tensor and the second tensor are respectively adjusted to the same number of channels as the input feature map by the Conv2d operation through the convolution layer, and then output after normalization and weighting processing through the normalization layer.

[0055] This embodiment improves the feature expression capability of the convolutional neural network model by combining the coordinate attention module and effectively introducing channel information into the attention mechanism.

[0056] For example, Figure 4 , which is the architecture diagram of the coordinate attention module of this embodiment. Wherein, H, W, C represent the height, width and number of channels of the input feature map respectively, and r represents the reduction rate.

[0057] The coordinate attention module in this embodiment is divided into two parts: information embedding and attention generation. Among them, the information is embedded in the structure through the global average pooling operation. Since ordinary average pooling compresses global information together and cannot obtain position information, this embodiment uses the global average pooling operation for information embedding, and by processing features in both width and height directions, the attention module can more accurately locate the target of interest in the model. The coordinate attention mechanism is decomposed by the following formula:

[0058]

[0059] Among them, Z c is the output associated with the cth channel; H and W represent the input feature map X c In this embodiment, a convolution kernel with dimensions (H, 1) or (1, W) is used to calculate all channels in the horizontal and vertical coordinates, respectively.

[0060] Furthermore, in an optional embodiment, the target detection model selects CIoULoss (Complete Intersection over Union Loss) as the loss function of the regression frame during its training process. Its expression is:

[0061]

[0062] CIoU Loss = 1–CIoU

[0063] Where IoU represents the intersection over union ratio, which is used to evaluate the overlap between the predicted value and the true value of the output box; ρ represents the Euclidean distance, b and b gt Represents the predicted box B and the real box B respectively gt The center point of the predicted box B and the real box B is represented by c. gt The diagonal length of the minimum bounding box; α is a positive trade-off parameter; v represents the consistency of the aspect ratio; Represents the true box B gt The aspect ratio, Represents the aspect ratio of the predicted box B.

[0064] In this embodiment, CIoU Loss is selected as the loss function of the regression frame, and the center point is used to normalize the distance. Among them, CIoU Loss is an improved version of GIoU Loss. CIoU Loss comprehensively considers multiple factors such as position, shape and direction, so that the model can learn the characteristics of the target frame more comprehensively. It is more sensitive to the position prediction of the target frame, which helps to improve the positioning accuracy.

[0065] Further optionally, the target detection model is pre-trained using an existing data set, where the data set mainly consists of visible light monitoring images and laser ranging data of marine vessels.

[0066] In a specific implementation process, the selected data set contains 7000 pictures, each with a resolution of 1920×1080, including images of ships of the categories of ore carriers, container ships, bulk carriers, general cargo ships, fishing boats and passenger ships, and the data set covers ship targets of various sizes, brightness and angles. Further, the data set is divided into training set, validation set and test set in a ratio of 8:1:1, and the improved YOLO detection method is trained. The training parameters are shown in Table 1 below.

[0067] Table 1 YOLO training parameters

[0068] parameter Parameter Value lr0 0.01 momentum 0.929 weight_decay 0.00005 epoch 100 batch_size 16 image_size 640

[0069] After the training is completed, the traditional YOLOv3, YOLOv5 and the improved YOLOv5 of this embodiment are compared. The specific comparison results are shown in Table 2 below. The target detection model described in this embodiment has a greater performance improvement.

[0070] Table 2 Model test accuracy

[0071] Model YOLOv3 YOLOv5 This embodiment Accuracy 0.667 0.718 0.892

[0072] Finally, the trained model is deployed to the optoelectronic pod to detect the ship and store the detected data in the memory.

[0073] In an optional embodiment, the distance direction data includes the laser beam position, laser beam direction, laser reflection distance and reflection intensity.

[0074] Then, in step S3, the step of generating a three-dimensional point cloud of the ship according to the distance direction data includes:

[0075] S3.1, preprocessing the distance direction data, including noise removal, error correction and filtering;

[0076] S3.2. Convert the preprocessed data into a point cloud format to obtain a three-dimensional point cloud of the ship.

[0077] Exemplarily, in step S3.1 of this embodiment, Gaussian filtering is optionally used to remove noise from the raw data; and, optionally, error correction is performed on the data based on the calibration parameters of the laser scanning device to eliminate the errors of the scanning instrument; and, optionally, points that are too far or too close, and invalid points (such as points beyond the scanning range) are filtered out to retain only valid data points.

[0078] Exemplarily, in step S3.2 of this embodiment, a laser beam is emitted by a laser ranging sensor on the optoelectronic pod device, and a light signal returned by a target ship at sea is received to measure the distance to the surface of the target ship, thereby generating high-precision three-dimensional point cloud data.

[0079] Exemplarily, in step S3.2 of this embodiment, the target ship at sea can also be photographed from different angles by a binocular camera carried on the optoelectronic pod device, and coordinate transformation can be performed based on the external parameter calibration matrix of the binocular camera to obtain three-dimensional point cloud data of the target ship.

[0080] In an optional embodiment, in step S4, the vessel image, the vessel three-dimensional point cloud and the vessel identification frame are fused based on a spatial transformation matrix, including the following steps:

[0081] S4.1. Constructing a spatial transformation matrix according to the parameters of the image acquisition sensor and the laser ranging sensor carried by the optoelectronic pod device;

[0082] S4.2, extracting feature points from the ship image and the three-dimensional point cloud based on the Canny algorithm and the improved Douglas-Peucker algorithm, calculating the spatial straight line intersections through the RANSAC algorithm, and obtaining a first fused image in combination with the spatial transformation matrix transformation;

[0083] S4.3. Use R-Tree spatial index algorithm to match the vessel identification frame with the first fused image to obtain a second fused image.

[0084] Exemplarily, in step S4.1, the positions of the image acquisition sensor and the laser ranging sensor are calculated, the camera internal and external parameter matrices are calculated, and the spatial transformation matrix is ​​obtained to map the point cloud data of each frame to the corresponding frame image.

[0085] The expression of the space transformation matrix is:

[0086]

[0087] In the formula, Z c is an axis parallel to the optical axis in the camera coordinate system; (u, v) is the pixel coordinate point of the two-dimensional image; f x 、f y are the horizontal and vertical focal lengths of the camera respectively, (u0, v0) is the optical center, and these four parameters constitute the camera intrinsic parameter matrix; [R|T] is the rotation and translation matrix, which constitutes the extrinsic parameter matrix; X ω , Y ω , Z ω represents the world coordinate system, which is defined as the radar coordinate system in this embodiment.

[0088] Exemplarily, in step S4.2, feature points are extracted from the optical image using the Canny algorithm and the improved Douglas-Peucker algorithm, the calibration plate plane is segmented in the scene using the RANSAC algorithm and the plane normal vector is calculated, and the boundary contour of the edge points of the plane plate is extracted; then the RANSAC algorithm is used to fit the spatial straight line of the plane calibration plate contour, two points are randomly selected from the target point cloud to generate a straight line, the distance between the remaining point cloud and the straight line is calculated, and the points less than the set threshold are regarded as internal points, and the process is repeated and the least squares method is used to fit the point set with the largest number of internal points as a straight line, and the equation of the spatial straight line is obtained, and then the feature points of the plane calibration plate are obtained, and then the intersection of the spatial straight line is calculated, and the camera internal and external parameter matrix parameters are substituted into the collinear equation, and the pixel coordinates of the corresponding points of the three-dimensional point cloud in the optical image can be obtained. Then, the RGB information of the corresponding pixel is assigned to the point cloud to achieve the fusion of the image and the point cloud, that is, the first fused image is obtained.

[0089] The RANSAC-based spatial straight line fitting algorithm used in this embodiment has the characteristics of high stability and can eliminate outliers.

[0090] The R-Tree used in step S4.3 of this embodiment is a dynamic balanced tree that organizes data according to the spatial position relationship of objects. It has strong flexibility and adjustability. The basic idea of ​​this tree structure is to divide the spatial object into a series of smaller rectangular areas. Each node represents a rectangular area, and the leaf node contains the actual spatial object. The rectangular area in the node is usually represented by a minimum bounding rectangle (MBR), which surrounds the minimum range of all objects in the node.

[0091] This embodiment uses the R-Tree algorithm to establish a data structure to describe the positional relationship between the detection box and the point cloud. Further, optionally, a two-layer R-Tree data structure is designed, wherein the first layer of the R-Tree stores the object detection box and the second layer stores the point cloud data points.

[0092] like Figure 5 , which is a schematic diagram of the R-Tree data structure of this embodiment, wherein R1 is the detection box of the detected target ship, and R2, R3, R4, ..., R7 are point cloud data points.

[0093] Furthermore, in an optional embodiment, the deep neural network includes 3 sequentially connected and hidden neural network layers; each of the neural network layers includes at least 256 neurons, and the activation function of each of the neural network layers includes a rectified linear function; and the last neural network layer is superimposed with 2 neurons and 1 sigmoid function.

[0094] The input of the deep neural network in this embodiment includes fused images and sensor attitude data, wherein the fused image provides the main and direct information of the ship image, and the sensor attitude data provides the attitude information of the ship.

[0095] Further optionally, the mean square error (MSE) is used as a loss function to measure the difference between the model output and the actual length, the deep neural network is trained, and the model parameters are adjusted using stochastic gradient descent (SGD) to minimize the loss function. The number of hidden layers is determined to be 3, the number of neurons is 256, and the learning rate is selected to be 0.01.

[0096] The output of the model in this embodiment is converted into a value in the range of 0-1 by a sigmoid function, which represents the estimated value of the length of the ship. This estimated length value can be used directly or further processed and interpreted according to specific application requirements. Figure 6 FIG. 1 is a schematic diagram of the measurement result of the length of a ship at sea according to the present embodiment.

[0097] Further optionally, an independent test dataset is used to evaluate and verify the trained model to check its performance on unseen data, and model parameters are adjusted or the model structure is improved based on the evaluation results to further improve the performance and generalization ability of the model, and the trained model is deployed in an optoelectronic pod to achieve real-time estimation of the ship's length.

[0098] Example 2

[0099] This embodiment applies the method for measuring the length of a ship at sea proposed in Embodiment 1, and proposes a system for measuring the length of a ship at sea with intelligent reconnaissance equipment, such as Figure 7 FIG. 1 is a schematic diagram of the marine vessel length measurement system according to the present embodiment.

[0100] The marine vessel length measurement system of the intelligent reconnaissance equipment proposed in this embodiment includes:

[0101] Intelligent reconnaissance equipment, including optoelectronic pod equipment and laser rangefinder sensors, used to collect ship images, distance direction data and sensor attitude data of target ships at sea;

[0102] An object detection module, which is equipped with a trained object detection model and is used to obtain a ship recognition frame based on an input ship image;

[0103] A point cloud generation module, used for generating a three-dimensional point cloud of a ship according to the distance direction data;

[0104] A fusion module, used for fusing the ship image, the ship three-dimensional point cloud and the ship identification frame based on a spatial transformation matrix to obtain a fused image;

[0105] A measurement module is equipped with a trained deep neural network and is used to output a ship length measurement result based on the input fused image and the sensor posture data.

[0106] It can be understood that the system of this embodiment corresponds to the method of the above-mentioned embodiment 1, and the options in the above-mentioned embodiment 1 are also applicable to this embodiment, so they will not be described repeatedly here.

[0107] Example 3

[0108] This embodiment proposes a computer device, including a memory and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor executes the steps of the marine vessel length measurement method proposed in Embodiment 1.

[0109] Example 4

[0110] This embodiment provides a storage medium having computer-readable instructions stored thereon, wherein the computer-readable instructions, when executed by a processor, implement the steps of the method for measuring the length of a marine vessel provided in Embodiment 1.

[0111] Exemplarily, the storage medium includes, but is not limited to, a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0112] Exemplarily, the instructions, programs, code sets or instruction sets may be implemented using conventional programming languages.

[0113] Exemplarily, the processor includes but is not limited to a smart phone, a personal computer, a server, a network device, etc., and is used to execute all or part of the steps of the offshore vessel length measurement method described in Example 1.

[0114] The terms in the drawings are only used for illustrative purposes and are not to be construed as limiting the present invention;

[0115] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A method for measuring the length of a marine vessel using intelligent reconnaissance equipment, characterized in that: The following steps are involved: The image acquisition sensor and laser ranging sensor on the optoelectronic pod device collect the ship image, distance direction data and sensor attitude data of the target ship at sea; Inputting the ship image into a trained object detection model to obtain a ship recognition frame; generating a three-dimensional point cloud of the vessel according to the distance direction data; The ship image, the ship three-dimensional point cloud and the ship identification frame are fused based on a spatial transformation matrix to obtain a fused image; The fused image and the sensor posture data are input into a trained deep neural network to obtain a ship length measurement result.

2. The method for measuring the length of a marine vessel according to claim 1, characterized in that: The target detection model includes a YOLOv5 network using mixed receptive field convolution; wherein the expression of the t-th group output of the mixed receptive field convolution is: In the formula, represents the feature map of the t-th group output, represents the feature map of the tth group of input; h, w, c represent the height, width and channel of the input feature map respectively; represents the convolution kernel of the tth group; k t represents the convolution kernel size of the tth group; m is the multiplication factor of the channel, c t Indicates the number of channels.

3. The method for measuring the length of a marine vessel according to claim 2, characterized in that: The target detection model also includes a coordinate attention module; the coordinate attention module includes a residual network layer, an average pooling layer, a compression layer, a spatial information encoding layer, a convolution layer and a normalization layer connected in sequence; Among them, after the input feature map is processed by the residual network layer, it enters the average pooling layer for average pooling in the horizontal and vertical directions respectively, to obtain a first one-dimensional vector and a second one-dimensional vector; the first one-dimensional vector and the second one-dimensional vector enter the compression layer for Concat and Conv2d operations to compress the number of channels, and then the spatial information in the vertical and horizontal directions is encoded by batch normalization and non-linear methods through the spatial information encoding layer, and is divided into a first tensor and a second tensor in the horizontal and vertical directions in the spatial dimension; the first tensor and the second tensor are respectively adjusted to the same number of channels as the input feature map by the Conv2d operation through the convolution layer, and then output after normalization and weighting processing through the normalization layer.

4. The method for measuring the length of a marine vessel according to claim 2, characterized in that: During the training process of the target detection model, CIoU Loss is selected as the loss function of the regression box; its expression is: CIoULoss=1–CIoU Where IoU represents the intersection over union ratio, which is used to evaluate the overlap between the predicted value and the true value of the output box; ρ represents the Euclidean distance, b and b gt Represents the predicted box B and the real box B respectively gt The center point of the predicted box B and the real box B is represented by c. gt The diagonal length of the minimum bounding box; α is a positive trade-off parameter; v represents the consistency of the aspect ratio; Represents the true box B gt The aspect ratio, Represents the aspect ratio of the predicted box B.

5. The method for measuring the length of a marine vessel according to claim 1, characterized in that: The distance direction data includes the laser beam position, laser beam direction, laser reflection distance and reflection intensity; the step of generating a three-dimensional point cloud of a ship according to the distance direction data includes: Preprocessing the distance direction data, including noise removal, error correction and filtering; The preprocessed data is converted into point cloud form to obtain the three-dimensional point cloud of the ship.

6. The method for measuring the length of a marine vessel according to any one of claims 1 to 5, characterized in that: The step of fusing the vessel image, the vessel three-dimensional point cloud and the vessel identification frame based on the spatial transformation matrix comprises the following steps: Constructing a spatial transformation matrix according to the parameters of the image acquisition sensor and the laser ranging sensor carried by the optoelectronic pod device; Extracting feature points from the ship image and the three-dimensional point cloud based on the Canny algorithm and the improved Douglas-Peucker algorithm, calculating spatial straight line intersections through the RANSAC algorithm, and obtaining a first fused image in combination with the spatial transformation matrix transformation; The vessel identification frame is matched with the first fused image using an R-Tree spatial index algorithm to obtain a second fused image.

7. The method for measuring the length of a marine vessel according to claim 6, characterized in that: The deep neural network includes three sequentially connected and hidden neural network layers; each of the neural network layers includes at least 256 neurons, and the activation function of each of the neural network layers includes a corrected linear function; and the last neural network layer is superimposed with two neurons and a sigmoid function.

8. A marine vessel length measurement system of intelligent reconnaissance equipment, using the marine vessel length measurement method according to any one of claims 1 to 7, characterized in that: include: Intelligent reconnaissance equipment, including optoelectronic pod equipment and laser rangefinder sensors, used to collect ship images, distance direction data and sensor attitude data of target ships at sea; An object detection module, which is equipped with a trained object detection model and is used to obtain a ship recognition frame based on an input ship image; A point cloud generation module, used for generating a three-dimensional point cloud of a ship according to the distance direction data; A fusion module, used for fusing the ship image, the ship three-dimensional point cloud and the ship identification frame based on a spatial transformation matrix to obtain a fused image; A measurement module is equipped with a trained deep neural network and is used to output a ship length measurement result based on the input fused image and the sensor posture data.

9. A device comprising a memory and a processor, wherein the memory stores computer-readable instructions, characterized in that: When the computer readable instructions are executed by the processor, the processor is caused to perform the steps of the marine vessel length measurement method according to any one of claims 1 to 7.

10. A storage medium having computer-readable instructions stored thereon, characterized in that: When the computer-readable instructions are executed by a processor, the steps of the method for measuring the length of a marine vessel according to any one of claims 1 to 7 are implemented.

Images