A method, device, equipment and medium for reconstructing a three-dimensional model of dynamic fish in a fish pass

Abstract

This application discloses a method, apparatus, equipment, and medium for reconstructing a three-dimensional model of dynamic fish within a fish passage, relating to the field of river ecological monitoring technology. The method achieves 1:1 reconstruction of real fish data based on depth image technology, and transforms the depth image data of swimming fish into linear state data using a deformation correction method (i.e., transforming dynamic fish into linear stationary fish), thereby completing the three-dimensional reconstruction. Based on this reconstructed fish model, complete and accurate fish body geometric parameters can be obtained, allowing for further analysis and extraction of fish features. This application improves the accuracy of body size measurement and feature extraction.

Description

Technical Field

[0001] This application relates to the field of river ecological monitoring technology, and in particular to a method, device, equipment and medium for reconstructing a three-dimensional model of dynamic fish in a fish passage. Background Technology

[0002] River damming and development have certain impacts on hydrological conditions, hydraulic characteristics, water temperature, water quality, and fish migration. According to the environmental protection requirements for water conservancy and hydropower projects, it is necessary to establish environmental protection measures to mitigate the impacts of dam construction, implement ecological compensation, and restore the health of the river ecosystem. Fish are relatively sensitive indicator species of river ecosystems, and changes in fish populations directly reflect the state of the river ecosystem. In recent years, water conservancy and hydropower projects have gradually begun to construct fish passage facilities, including fishways, fish ladders, fish lifts, and fish transport vessels, as connecting channels between upstream and downstream of the dam. The effective operation of fish passage facilities is an important means of protecting the longitudinal connectivity of rivers, playing a significant role in mitigating the obstructive effects of dams, promoting fish migration, and facilitating upstream and downstream population and gene exchange.

[0003] Monitoring the effectiveness of fish passage facilities is a necessary step in their operation. By monitoring the actual operational effectiveness of these facilities and analyzing the movement behavior and changes of fish as they pass through, problems in the design and construction process can be identified. This provides a basis for improving and optimizing the function of the fish passage facilities and accumulates valuable experience. Monitoring the operational effectiveness of fish passage facilities mainly involves recording several biological and abiotic indicators. Biological indicators include the number, species, body shape, age group, and quality of fish passing through, while abiotic indicators include water temperature, flow rate, head difference, and water level changes. Among these, monitoring biological indicators is both the key focus and the most challenging aspect of fish passage effectiveness monitoring.

[0004] In recent years, with the rapid development of machine vision technology, measuring the body size parameters of animals through non-contact methods has become an inevitable trend. However, since fish are in an active state during the monitoring process, and the visual technology images are also images of fish in various dynamic postures, the difficulty in measuring the body size and extracting features of fish has also affected the accuracy of body size measurement and feature extraction. Summary of the Invention

[0005] The purpose of this application is to provide a method, apparatus, equipment, and medium for reconstructing a three-dimensional model of dynamic fish in a fish passage, so as to improve the accuracy of body size measurement and feature extraction.

[0006] To achieve the above objectives, this application provides the following solution.

[0007] Firstly, this application provides depth image data for acquiring dynamic fish within a fish passage;

[0008] The depth image data is preprocessed to obtain preprocessed depth image data;

[0009] Intelligent fish segmentation is performed on the preprocessed depth image data to obtain fish outlines under dynamic conditions.

[0010] Deformation correction is performed on the fish outline map under dynamic conditions to obtain the fish outline coordinate data and fish skeleton coordinate data under straight line conditions.

[0011] Three-dimensional reconstruction is performed based on the fish outline coordinate data and fish skeleton coordinate data in a straight line state to obtain a three-dimensional reconstruction model of the fish in a straight line state.

[0012] Secondly, this application provides a three-dimensional model reconstruction device for dynamic fish within a fish passage. The device utilizes the aforementioned three-dimensional model reconstruction method for dynamic fish within a fish passage. The three-dimensional model reconstruction device for dynamic fish within a fish passage includes:

[0013] The depth image acquisition module is used to acquire depth image data of dynamic fish in the fish passage;

[0014] The preprocessing module is used to preprocess the depth image data to obtain preprocessed depth image data;

[0015] The fish intelligent segmentation module is used to perform intelligent fish segmentation on preprocessed depth image data to obtain fish outline maps under dynamic conditions.

[0016] The correction module is used to perform deformation correction on the fish outline map under dynamic conditions, and obtain the fish outline coordinate data and fish skeleton coordinate data under straight line conditions.

[0017] The 3D reconstruction module is used to perform 3D reconstruction based on the fish outline coordinate data and fish skeleton coordinate data in a straight line state, so as to obtain a 3D reconstruction model of the fish in a straight line state.

[0018] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described method for reconstructing a three-dimensional model of dynamic fish in a fish passage.

[0019] Fourthly, this application provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-described method for reconstructing a three-dimensional model of dynamic fish within a fish passage.

[0020] According to the specific embodiments provided in this application, this application has the following technical effects.

[0021] This application provides a method, apparatus, device, and medium for reconstructing a three-dimensional model of dynamic fish within a fish passage. Based on depth image technology, this application achieves a 1:1 reconstruction of real fish data. Using a deformation correction method, it transforms the swimming fish data into linear state data (i.e., converting dynamic fish into linear stationary fish), thereby completing the three-dimensional reconstruction. Based on this reconstructed fish model, complete and accurate fish body geometric parameters can be obtained, allowing for further analysis and extraction of fish features. This application improves the accuracy of body size measurement and feature extraction. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a flowchart illustrating a method for reconstructing a three-dimensional model of dynamic fish within a fish passage, as provided in an embodiment of this application.

[0024] Figure 2 This is a schematic diagram of the structure of a CNN network model provided in an embodiment of this application.

[0025] Figure 3 This is a schematic diagram of the structure of a fish segmentation network model provided in an embodiment of this application.

[0026] Figure 4 This is a flowchart of a deformation correction algorithm provided in an embodiment of this application.

[0027] Figure 5 This is a framework diagram of fish features provided in an embodiment of this application.

[0028] Figure 6 This is a schematic diagram of fish body partitioning provided in an embodiment of this application.

[0029] Figure 7 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0030] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0031] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0032] With the widespread application of depth cameras, acquiring 3D data of objects has become more convenient. Therefore, a 3D reconstruction and body size measurement scheme for underwater fish based on a dual-view depth camera was developed. This scheme can not only obtain the 3D contours of dynamic underwater fish, but also perform 1:1 3D reconstruction, thereby extracting the body size information of the fish and analyzing and identifying fish features through a realistic 3D reconstruction model.

[0033] In one exemplary embodiment, such as Figure 1 and Figure 2 As shown, a method for reconstructing a three-dimensional model of dynamic fish in a fish passage is provided, including the following steps 101 to 105.

[0034] Step 101: Obtain depth image data of dynamic fish within the fish passage.

[0035] Step 102: Preprocess the depth image data to obtain preprocessed depth image data.

[0036] Step 103: Perform intelligent fish segmentation on the preprocessed depth image data to obtain fish outline maps under dynamic conditions.

[0037] Step 104: Perform deformation correction on the fish outline map under dynamic conditions to obtain the fish outline coordinate data and fish skeleton coordinate data under straight line conditions.

[0038] Step 105: Based on the fish outline coordinate data and fish skeleton coordinate data in a straight line state, perform three-dimensional reconstruction to obtain a three-dimensional reconstruction model of the fish in a straight line state.

[0039] By implementing steps 101 to 105 above, a 3D reconstruction model of the fish in a 1:1 straight line state can be achieved, improving the accuracy of body size measurement and feature extraction.

[0040] In another exemplary embodiment of this application, in order to acquire depth image data of dynamic fish within a fish passage, a depth camera is installed on each of the left and right sides, top and bottom sides, and front and back sides of the measurement channel to collect depth image data of dynamic fish within the fish passage. In this step, depth image data (including phenotypic image data and depth information data, wherein the depth information represents the distance of the target object from the acquisition device) of dynamic fish within the fish passage is acquired through the three depth cameras from different perspectives.

[0041] In another exemplary embodiment of this application, step 102 preprocesses the acquired depth image data by using a CNN network model to perform denoising, dehazing, deblurring, and restoration operations on the underwater depth image data. Figure 2 As shown, this CNN network model is based on the CHAN framework and includes a generator and a discriminator. The generator further includes an encoding network G. Enc Decoding Network G Dec and Feature Reshaping Network G Rec The generator takes a distorted underwater image as input (i.e., depth image data acquired by a depth camera). First, the depth image data is downsampled using an encoding network. Then, a feature reshaping network extracts deeper features. Finally, a decoding network upsamples the extracted features to restore the original image size, resulting in an enhanced, clear underwater image (i.e., preprocessed depth image data). Further, the encoding network extracts features using a Feature Attention Residual (FAR) module, and the decoding network refines features using a Strengthen-Operate-Subtract (SOS) module. The entire generator is supplemented with an Adaptive Dense Feature Fusion (ADFF) module for feature fusion. The discriminator consists of multiple convolutional layers connected in sequence and a Multi-scale fusion (MF) module. This MF module is connected to the output of the last convolutional layer in the series of connected convolutional layers. The discriminator takes an enhanced, sharp underwater image and its corresponding ground truth image as input. In the discriminator, features are first extracted from the input image using convolutional layers, followed by multi-scale feature discrimination using a multi-scale feature extraction (MF) module, yielding a discriminant value used to supervise the generator's training. During training, the generator and discriminator are trained alternately to update gradient parameters. After training, only the trained generator is used for depth image data augmentation. Then, a semantic segmentation algorithm is used to extract fish image data.

[0042] In another exemplary embodiment of this application, step 103 preprocesses the acquired depth image data, and performs intelligent fish segmentation on the preprocessed depth image data based on a residual network and a multi-layer parallel dilated convolution fish segmentation network model to obtain a dynamic fish outline map.

[0043] Figure 3 (a) in the diagram is a schematic diagram of the fish segmentation network model. Figure 3 (b) in the diagram is a schematic diagram of the structure of the first upsampling module in the fish segmentation network model. Figure 3As shown, the fish segmentation network model includes a model encoder and a model decoder.

[0044] The model encoder includes: a first encoding module, a second encoding module, a third encoding module, a fourth encoding module, and a fifth encoding module connected in sequence; the first encoding module includes: a 7×7 convolutional layer, a regularization layer, and a ReLU layer connected in sequence; the second encoding module includes: 3 residual modules; the third encoding module includes: 4 residual modules; the fourth encoding module includes: 6 residual modules; the fifth encoding module includes: 3 residual modules; each residual module includes: a 1×1 convolutional layer, a 3×3 convolutional layer, a 1×1 convolutional layer, and a residual connection structure connected in sequence.

[0045] The encoding process of the model encoder is as follows: the preprocessed depth image data is subjected to 7×7 convolution, regularization, and ReLU to obtain the first encoding feature map. The first encoding feature map is passed through 3 residual modules to obtain the second encoding feature map. The second encoding feature map is passed through 4 residual modules to obtain the third encoding feature map. The third encoding feature map is passed through 6 residual modules to obtain the fourth encoding feature map. The fourth encoding feature map is passed through 3 residual modules to obtain the fifth encoding feature map.

[0046] The model encoder in this application uses a residual network as the backbone to extract target semantic information, which can effectively avoid gradient vanishing and gradient exploding while extracting features at a depth.

[0047] The model decoder includes: multiple first upsampling modules and second upsampling modules connected in sequence, wherein the second upsampling module is connected to the last first upsampling module among the multiple first upsampling modules connected in sequence; the first upsampling module includes a splicing and fusion layer, a 3×3 convolutional layer, a 3×3 convolutional layer and a multi-layer parallel dilated convolutional layer connected in sequence; the second upsampling module includes a splicing and fusion layer, a 3×3 convolutional layer, a 3×3 convolutional layer, a multi-layer parallel dilated convolutional layer and a 1×1 convolutional layer connected in sequence.

[0048] The decoding process of the model decoder is as follows: The feature map obtained by the model encoder is spliced ​​and fused with the high-level semantic information feature map generated by upsampling and the shallow semantic information feature map generated by the corresponding stage in the first half through a skip connection. The fused decoded feature map is then subjected to two 3×3 convolutions and a multi-layer parallel dilated convolution with four sampling rates of 1, 6, 12 and 18 to expand the receptive field and supplement the missing information after the fish edge convolution. Finally, a clear and accurate fish outline map is obtained through the classifier.

[0049] In another exemplary embodiment of this application, in step 104 above, dynamic fish contour coordinate data and dynamic fish skeleton coordinate data are obtained by edge detection and skeleton extraction using the Hough transform algorithm. Simultaneously, the fish contour image is corrected to determine the orientation information of the skeleton and the cutting points required for correction. Finally, the fish contour coordinate data and fish skeleton coordinate data in a straight line state are obtained through image cutting, rotation, and synthesis. Specifically, this step can be replaced by steps 201-204.

[0050] Step 201: Use an edge detection algorithm to perform edge detection on the dynamic fish outline map to obtain dynamic fish outline coordinate data and dynamic fish skeleton coordinate data; the dynamic fish outline coordinate data includes multiple dynamic fish outline coordinates, and the dynamic fish skeleton coordinate data includes multiple dynamic fish skeleton coordinates.

[0051] Step 202: Construct a central straight line based on the dynamic fish skeleton coordinate data; the number of dynamic fish skeleton coordinate points located on the central straight line in the dynamic fish skeleton coordinate data is greater than a preset threshold.

[0052] Step 203: Determine the deviation distance of each dynamic fish skeleton coordinate point relative to the central straight line.

[0053] Step 204: Based on the deviation distance of each dynamic fish skeleton coordinate point relative to the central straight line, offset each dynamic fish skeleton coordinate point and the corresponding dynamic fish outline coordinates to obtain fish outline coordinate data and fish skeleton coordinate data in a straight line state.

[0054] like Figure 4 As shown, the specific algorithm description for step 104 above is as follows:

[0055] Image preprocessing: Grayscale conversion, which converts color images into grayscale images to simplify the processing.

[0056] Edge detection: Using the Canny edge detector or other edge detection algorithms to identify edges in an image.

[0057] Parameter space initialization: Define the parameter space, ρ (the perpendicular distance from the origin to the line) and θ (the angle between the line and the x-axis).

[0058] Accumulator array creation: Set the accumulator size, determine the range of ρ and θ according to the required precision, and set the size of the accumulator array accordingly.

[0059] Mapping to Hough space: Map each edge point. For each edge point (x, y) in the image, calculate all possible line parameters (ρ, θ) that it might correspond to. For each possible (ρ, θ), increment the count at the corresponding position in the accumulator array.

[0060] Finding peaks: A threshold is set, and only (ρ, θ) pairs whose accumulator values ​​exceed this threshold are considered valid line parameters. Local maxima are searched in the accumulator array to avoid detecting multiple overlapping lines.

[0061] Extracting the equation of the line: For each peak location, obtain its ρ and θ values. Use the ρ and θ values ​​to calculate the standard equation or polar equation of the line.

[0062] Results visualization: Plot the detected lines on the original image or output them to a new image.

[0063] In another exemplary embodiment of this application, in step 105 above, a three-dimensional fish reconstruction model is obtained by performing three-dimensional modeling of the fish using the fish outline coordinate data and the fish skeleton coordinate data in a straight line state. Point cloud computing, point cloud registration, and data fusion are then performed using the fish outline coordinate data and fish skeleton coordinate data obtained in step 104 to finally obtain a refined three-dimensional fish reconstruction model in a straight line state and output the coordinate data of the three-dimensional fish reconstruction model.

[0064] The specific steps are as follows:

[0065] Data preprocessing: Remove noise points from point cloud data and align point cloud data acquired from different perspectives or at different times.

[0066] Data simplification: Reduce the amount of point cloud data to lower the computational complexity of subsequent processing, while preserving the geometric features of the model as much as possible. Convert the point cloud data into a triangular mesh model to facilitate subsequent modeling and visualization.

[0067] Feature extraction and surface reconstruction: Calculating the normal direction of each point helps in understanding the geometric features of the surface. Surface reconstruction algorithms (such as Poisson reconstruction, implicit surface reconstruction, etc.) are used to convert point cloud data into continuous surface models.

[0068] Model optimization: Analyze the errors in the reconstructed model and make necessary adjustments and optimizations. Add details to the model as needed to make it more realistic.

[0069] Model Export: Export the final model to a commonly used 3D file format.

[0070] In another exemplary embodiment of this application, after step 105 above, a step of calculating the body size parameters and extracting feature indicators of fish based on a three-dimensional reconstruction model of fish in a straight line state is also performed.

[0071] Body size parameters: Key body size parameters of the fish are calculated using coordinate data from the 3D reconstruction model of the fish in a straight line state. These body size parameters include body length, body width, and body height. The body length is calculated by the distance between the coordinates of the fish's lips and tail; the body width is calculated by the distance between the coordinates of the fish's two sides; and the body height is calculated by the distance between the coordinates of the fish's highest and lowest points.

[0072] Feature extraction: Shape and texture features are extracted from a 3D reconstruction model of a fish in a straight line, such as... Figure 5 As shown and Figure 6 As shown, morphological features such as the number, position, and shape of the dorsal fin, caudal fin, pectoral fin, and caudal fin, as well as the outline of the fish's eyes and spine, can be directly extracted from the image. These features are classified as primary indicators of fish phenotypic features. Features that require secondary calculation, such as the ratio of long to short axes, tail shape, fin shape, and relative position, can be classified as secondary indicators of fish phenotypic features. A fish phenotypic feature database is established using the obtained primary and secondary indicator parameters.

[0073] According to the specific embodiments provided in this application, this application has the following technical effects.

[0074] The depth image data of dynamic fish in the fish passage is obtained from three directions using a depth camera, and the image data is restored to a 1:1 scale based on the depth image data.

[0075] By using a deformation correction algorithm, swimming fish are transformed into a static state and their complete geometric parameters are obtained, which allows for further analysis and extraction of fish features.

[0076] Based on the same inventive concept, this application also provides a device for reconstructing a three-dimensional model of dynamic fish in a fish passage, which is used to implement the above-mentioned method for reconstructing a three-dimensional model of dynamic fish in a fish passage. The solution provided by this device is similar to the solution described in the above-described method. Therefore, the specific limitations of one or more embodiments of the device for reconstructing a three-dimensional model of dynamic fish in a fish passage provided below can be found in the limitations of the three-dimensional model reconstruction method for dynamic fish in a fish passage described above, and will not be repeated here.

[0077] In one exemplary embodiment, a three-dimensional model reconstruction device for dynamic fish within a fish passage is provided, comprising:

[0078] The depth image acquisition module is used to acquire depth image data of dynamic fish within the fish passage.

[0079] The preprocessing module is used to preprocess the depth image data to obtain preprocessed depth image data.

[0080] The fish intelligent segmentation module is used to perform intelligent fish segmentation on preprocessed depth image data to obtain fish outline maps under dynamic conditions.

[0081] The correction module is used to perform deformation correction on the fish outline map under dynamic conditions, and obtain the fish outline coordinate data and fish skeleton coordinate data under straight line conditions.

[0082] The 3D reconstruction module is used to perform 3D reconstruction based on the fish outline coordinate data and fish skeleton coordinate data in a straight line state, so as to obtain a 3D reconstruction model of the fish in a straight line state.

[0083] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 7 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database is used for reconstructing and measuring results. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for reconstructing a three-dimensional model of dynamic fish within a fish passage.

[0084] Those skilled in the art will understand that Figure 7 The structures shown are merely block diagrams of some structures related to the present application and do not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than shown in the figures, or combine certain components, or have different component arrangements. In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0085] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0086] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0087] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0088] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0089] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0090] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for reconstructing a three-dimensional model of dynamic fish in a fish passage, characterized by, include: Acquire depth image data of dynamic fish within the fish passage; The depth image data is preprocessed to obtain preprocessed depth image data; Intelligent fish segmentation is performed on the preprocessed depth image data to obtain fish outlines under dynamic conditions. Deformation correction is performed on the fish outline map under dynamic conditions to obtain the fish outline coordinate data and fish skeleton coordinate data under straight line conditions. Three-dimensional reconstruction is performed based on the fish outline coordinate data and fish skeleton coordinate data in a straight line state to obtain a three-dimensional reconstruction model of the fish in a straight line state. Intelligent fish segmentation is performed on the preprocessed depth image data to obtain dynamic fish outline maps, specifically including: A fish segmentation network model is used to perform intelligent fish segmentation on preprocessed depth image data to obtain fish outline maps under dynamic conditions. The fish segmentation network model includes: a model encoder and a model decoder; The model encoder includes: a first encoding module, a second encoding module, a third encoding module, a fourth encoding module, and a fifth encoding module connected in sequence; The first encoding module includes: a 7×7 convolutional layer, a regularization layer, and a ReLU layer connected in sequence; The second encoding module includes: 3 residual modules; The third encoding module includes: 4 residual modules; The fourth encoding module includes: 6 residual modules; The fifth encoding module includes: 3 residual modules; The residual module includes: a 1×1 convolutional layer, a 3×3 convolutional layer, a 1×1 convolutional layer, and a residual connection structure connected in sequence; The model decoder includes: a second upsampling module and a plurality of first upsampling modules connected in sequence, wherein the second upsampling module is connected to the last of the plurality of first upsampling modules connected in sequence. The first upsampling module includes a stitching and fusion layer, a 3×3 convolutional layer, a 3×3 convolutional layer and a multi-layer parallel dilated convolutional layer connected in sequence; The second upsampling module includes a stitching and fusion layer, a 3×3 convolutional layer, a 3×3 convolutional layer, a multi-layer parallel dilated convolutional layer, and a 1×1 convolutional layer connected in sequence.

2. The method for reconstructing a three-dimensional model of dynamic fish within a fish passage according to claim 1, characterized in that, The depth image data is preprocessed to obtain preprocessed depth image data, specifically including: A generator is used to downsample, extract deep features, and upsample the depth image data to obtain preprocessed depth image data.

3. The method for reconstructing a three-dimensional model of dynamic fish within a fish passage according to claim 2, characterized in that, The generator comprises an encoding network, a feature reshaping network, and a decoding network connected in sequence. The encoding network, the feature reshaping network, and the decoding network are used for downsampling, deep feature extraction, and upsampling, respectively.

4. The method for reconstructing a three-dimensional model of dynamic fish within a fish passage according to claim 2, characterized in that, The generator is obtained through iterative training with the discriminator. The discriminator includes multiple convolutional layers and an MF module connected in sequence. The MF module is connected to the output of the last convolutional layer in the multiple convolutional layers connected in sequence. The MF module is a multi-scale fusion module.

5. The method for reconstructing a three-dimensional model of dynamic fish within a fish passage according to claim 1, characterized in that, Deformation correction is performed on the fish outline image under dynamic conditions to obtain the fish outline coordinate data and fish skeleton coordinate data under straight-line conditions. Specifically, this includes: An edge detection algorithm is used to perform edge detection on the dynamic fish outline map to obtain dynamic fish outline coordinate data and dynamic fish skeleton coordinate data; the dynamic fish outline coordinate data includes multiple dynamic fish outline coordinates, and the dynamic fish skeleton coordinate data includes multiple dynamic fish skeleton coordinates. A central straight line is constructed based on dynamic fish skeleton coordinate data; the number of dynamic fish skeleton coordinate points located on the central straight line in the dynamic fish skeleton coordinate data is greater than a preset threshold. Determine the deviation distance of each dynamic fish skeleton coordinate point relative to the central straight line; Based on the deviation distance of each dynamic fish skeleton coordinate point relative to the central straight line, the coordinates of each dynamic fish skeleton and the corresponding dynamic fish outline coordinates are offset to obtain the fish outline coordinate data and fish skeleton coordinate data in the straight line state.

6. The method for reconstructing a three-dimensional model of dynamic fish within a fish passage according to claim 1, characterized in that, Based on the fish outline coordinate data and fish skeleton coordinate data in a straight line state, a 3D reconstruction is performed to obtain a 3D reconstructed model of the fish in a straight line state. This is followed by: The body size parameters of fish and the feature indicators are calculated based on the three-dimensional reconstruction model of fish in a straight line state. The body size parameters include body length, body width and body height. The feature indicators include primary indicators and secondary indicators. The primary indicators include color, shape and texture. The secondary indicators include the ratio of major axis to minor axis, tail shape features, fin shape and relative position.

7. A three-dimensional model reconstruction device for dynamic fish in a fish passage, characterized in that, The three-dimensional model reconstruction device for dynamic fish in the fish passage uses the three-dimensional model reconstruction method for dynamic fish in the fish passage according to any one of claims 1-6, and the three-dimensional model reconstruction device for dynamic fish in the fish passage includes: The depth image acquisition module is used to acquire depth image data of dynamic fish in the fish passage; The preprocessing module is used to preprocess the depth image data to obtain preprocessed depth image data; The fish intelligent segmentation module is used to perform intelligent fish segmentation on preprocessed depth image data to obtain fish outline maps under dynamic conditions. The correction module is used to perform deformation correction on the fish outline map under dynamic conditions, and obtain the fish outline coordinate data and fish skeleton coordinate data under straight line conditions. The 3D reconstruction module is used to perform 3D reconstruction based on the fish outline coordinate data and fish skeleton coordinate data in a straight line state, so as to obtain a 3D reconstruction model of the fish in a straight line state.

8. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the method for reconstructing a three-dimensional model of dynamic fish within a fish passage as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the method for reconstructing a three-dimensional model of dynamic fish within a fish passage as described in any one of claims 1-6.

Images