A modeling method and a modeling system for a bicycle shock absorber

Abstract

The present invention discloses a modeling method and a modeling system for a bicycle shock absorber, belonging to the technical field of image modeling, including: acquiring shock absorber images of different types and different working states of the bicycle shock absorber; annotating key parts of the shock absorber images, processing the annotated shock absorber images to obtain a preprocessed image dataset; constructing a force point activation analysis model based on a convolutional neural network, inputting the preprocessed image dataset into the force point activation analysis model for calculation to obtain a shock absorber force high-activation region vector; constructing a three-dimensional model of the bicycle shock absorber based on the shock absorber force high-activation region vector and shock absorber parameters. The force point vector obtained by the present invention through CNN can provide accurate position information, which helps to accurately understand the force distribution of the shock absorber under different conditions.

Description

Technical Field

[0001] The present invention belongs to the technical field of image modeling, and particularly relates to a modeling method and a modeling system for a bicycle shock absorber. Background Art

[0002] Bicycle shock absorbers are mainly used to absorb the impacts and vibrations caused by uneven road surfaces, and improve the riding comfort and controllability. Common shock absorbers include: Spring shock absorbers: Using a combination of a metal spring and a damper, and the damper is used to control the rebound of the spring. Gas shock absorbers: Utilizing the compression properties of gas and oil to absorb impacts.

[0003] Currently, the modeling of bicycle shock absorbers mainly inputs the morphological parameters and mechanical parameters of the shock absorbers into software for modeling. It is impossible to accurately locate and identify the force application points of the bicycle shock absorbers. Since standard parameter modeling is used in the modeling process, it is impossible to adjust the response of the shock absorbers during the manufacturing process to adapt to different road conditions. Summary of the Invention

[0004] To solve the above technical problems, the present invention proposes a modeling method and a modeling system for a bicycle shock absorber to solve the problems existing in the above prior art.

[0005] To achieve the above object, the present invention provides a modeling method for a bicycle shock absorber, including:

[0006] Obtaining shock absorber images of different types and different working states of the bicycle shock absorber;

[0007] Annotating key parts of the shock absorber images, processing the annotated shock absorber images, and obtaining a preprocessed image dataset;

[0008] Constructing a force application point activation analysis model based on a convolutional neural network, inputting the preprocessed image dataset into the force application point activation analysis model for calculation, and obtaining a high-activation region vector of the shock absorber force;

[0009] Constructing a three-dimensional model of the bicycle shock absorber based on the high-activation region vector of the shock absorber force and shock absorber parameters.

[0010] Preferably, the process of obtaining shock absorber images of different types and different working states of the bicycle shock absorber includes:

[0011] Shooting different working states of different types of bicycle shock absorbers based on a depth camera to obtain shooting images;

[0012] The different types of bicycle shock absorbers include but are not limited to spring shock absorbers and gas shock absorbers;

[0013] The different working states include, but are not limited to, a compressed state and an uncompressed state.

[0014] Preferably, the process of annotating the key parts of the shock absorber image includes:

[0015] Obtain the coordinates of the force application points of the shock absorber in the shock absorber image, and generate shock absorber force application point coordinates;

[0016] Convert the shock absorber force application point coordinates into force application point vectors, and decompose the force application point vectors into components along the working direction and the perpendicular direction of the shock absorber;

[0017] Annotate the components along the working direction and the perpendicular direction of the shock absorber to the shock absorber image, and complete the annotation of the key parts of the shock absorber image.

[0018] Preferably, the process of processing the annotated shock absorber image to obtain a preprocessed image dataset includes:

[0019] Obtain the morphological characteristics of the shock absorber, and generate a pixel point threshold range based on the morphological characteristics of the shock absorber;

[0020] Eliminate the pixel points in each pixel point of the annotated shock absorber image that do not conform to the pixel point threshold range, and generate a shock absorber feature image;

[0021] Unify the sizes of all shock absorber feature images to generate the preprocessed image dataset.

[0022] Preferably, the process of constructing a force application point activation analysis model based on a convolutional neural network includes:

[0023] Obtain a sample set of bicycle shock absorbers, and divide the sample set into a training set and a test set;

[0024] Construct an attention mechanism module based on the Transformer module;

[0025] Construct a convolutional neural network, and import the attention mechanism module into the convolutional neural network to generate an improved convolutional neural network;

[0026] Train the improved convolutional neural network with the training set to obtain a training model;

[0027] Train the training model with the test set to generate a training result;

[0028] Fine-tune the parameters of the training model based on the training result to generate the force application point activation analysis model;

[0029] The sample set of the bicycle shock absorber includes force application point analysis information.

[0030] Preferably, the process of inputting the preprocessed image dataset into the force application point activation analysis model for calculation to obtain the high-activation region vector of the shock absorber force includes:

[0031] Extracting the local force characteristics of the preprocessed image dataset through the attention mechanism module of the force application point activation analysis model to obtain the force attention score;

[0032] Obtaining the force application point distribution vector based on the force attention score and the convolution module of the force application point activation analysis model;

[0033] Calculating the correlation degree of the force application point distribution vector, and partitioning the force application point distribution vector based on the correlation degree to obtain the high-activation region vector of the shock absorber force.

[0034] Preferably, the expression for obtaining the force attention score is:

[0035]

[0036] where Score(f x , f y ) represents the force attention score, f x represents the component in the working direction of the shock absorber, f y represents the component in the vertical direction of the shock absorber, T represents matrix transpose, represents the square root of the dimension of the component in the vertical direction of the shock absorber.

[0037] Preferably, the process of constructing a three-dimensional model of a bicycle shock absorber based on the high-activation region vector of the shock absorber force and shock absorber parameters includes:

[0038] Constructing a shock absorber model based on the shock absorber parameters;

[0039] Importing the high-activation region vector of the shock absorber force as the force attribute of the shock absorber into the shock absorber model to generate a shock absorber force model;

[0040] Verifying the structural strength and function of the shock absorber force model through the finite element analysis method to obtain the three-dimensional model of the bicycle shock absorber.

[0041] To achieve the above object, the present invention also provides a modeling system for a bicycle shock absorber, including:

[0042] A data acquisition subsystem for acquiring shock absorber images of different types and different working states of a bicycle shock absorber;

[0043] A data processing subsystem for annotating key parts of the shock absorber image, processing the annotated shock absorber image to obtain a preprocessed image dataset;

[0044] A calculation subsystem for constructing a force application point activation analysis model based on a convolutional neural network, inputting the preprocessed image dataset into the force application point activation analysis model for calculation to obtain a shock absorber force high-activation region vector;

[0045] A modeling subsystem for constructing a three-dimensional model of a bicycle shock absorber based on the shock absorber force high-activation region vector and shock absorber parameters.

[0046] Compared with the prior art, the present invention has the following advantages and technical effects:

[0047] The force point vector obtained by the present invention through CNN can provide accurate position information, which helps to accurately understand the force distribution of the shock absorber under different conditions, can better analyze the stress and strain of the shock absorber structure in the three-dimensional model of the shock absorber, calculate the stress distribution under various loading conditions, effectively avoid over-design or material waste, and at the same time ensure the reliability and durability of the shock absorber. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The accompanying drawings constituting a part of this application are used to provide a further understanding of this application. The schematic embodiments and descriptions thereof of this application are used to explain this application and do not constitute an improper limitation to this application. In the drawings:

[0049] Figure 1 is a flowchart of a method for modeling a bicycle shock absorber according to an embodiment of the present invention;

[0050] Figure 2 is a schematic diagram of a modeling system of a bicycle shock absorber according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0051] It should be noted that, without conflict, the embodiments in this application and the features in the embodiments can be combined with each other. The following will refer to the accompanying drawings and combine the embodiments to detail this application.

[0052] It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0053] Embodiment 1

[0054] As Figure 1 shown, this embodiment provides a method for modeling a bicycle shock absorber, including:

[0055] Obtain shock absorber images of different types and different working states of bicycle shock absorbers;

[0056] Annotate the key parts of the shock absorber images, and process the annotated shock absorber images to obtain a preprocessed image dataset;

[0057] Build a force application point activation analysis model based on a convolutional neural network, input the preprocessed image dataset into the force application point activation analysis model for calculation, and obtain a high-activation region vector of the shock absorber force;

[0058] Build a 3D model of the bicycle shock absorber based on the high-activation region vector of the shock absorber force and shock absorber parameters.

[0059] For a further optimized solution, the process of obtaining shock absorber images of different types and different working states of bicycle shock absorbers includes:

[0060] Based on a depth camera, photograph different working states of different types of bicycle shock absorbers to obtain photographed images; in this embodiment, images of various bicycle shock absorbers are collected. These images can include shock absorbers of different brands, different types (such as front forks, rear shock absorbers) and different working states (compressed and uncompressed states), and are obtained using a depth camera; annotate the collected images to mark the key parts of the shock absorber, such as springs, dampers, locking mechanisms, etc. The annotation can be done manually or assisted by semi-automatic tools.

[0061] The different types of bicycle shock absorbers include, but are not limited to, spring shock absorbers and gas shock absorbers;

[0062] The different working states include, but are not limited to, the compressed state and the uncompressed state.

[0063] For a further optimized solution, the process of annotating the key parts of the shock absorber images includes:

[0064] Obtain the coordinates of the force application points of the shock absorber in the shock absorber image to generate shock absorber force application point coordinates;

[0065] Convert the shock absorber force application point coordinates into a force application point vector, and decompose the force application point vector into components along the working direction and the vertical direction of the shock absorber;

[0066] Annotate the components along the working direction and the vertical direction of the shock absorber to the shock absorber image to complete the annotation of the key parts of the shock absorber image.

[0067] For a further optimized solution, the process of processing the annotated shock absorber images to obtain a preprocessed image dataset includes:

[0068] Obtain the morphological features of the shock absorber, and generate a pixel threshold range based on the morphological features of the shock absorber;

[0069] Eliminate the pixels in each pixel of the labeled shock absorber image that do not conform to the pixel threshold range to generate a shock absorber feature image;

[0070] Unify the sizes of all shock absorber feature images to generate the preprocessed image dataset, and the image data can be enhanced by methods such as rotation, scaling, and color adjustment to improve the generalization ability of the model.

[0071] Perform Mosaic image data augmentation on all shock absorber feature images, randomly read four original image data, randomly select a region as the central region for splicing, and splice the four images that have been randomly cropped, randomly scaled, and randomly arranged according to the central region to obtain the preprocessed image;

[0072] Perform Letterbox image data augmentation on the image data after Mosaic image data augmentation, scale the long side of the image data to a specified size, and fill the short side with black pixels to obtain the preprocessed image. The formula for Letterbox image data augmentation is as follows:

[0073]

[0074] Among them, d is the length to be filled on one side, w is the width of the original image, h is the height of the original image, w′ is the specified width after image scaling, h′ is the specified height after image scaling. The improved YOLOv8 forest fire detection model is downsampled 5 times in total, and the image features are scaled to 1 / 32 of the input image. Therefore, the size of the input image needs to be a multiple of 32. Through the mod32 operation, the total filling length is obtained, and / 2 is used to obtain the filling length on one side.

[0075] For a further optimization plan, the process of constructing a focus activation analysis model based on a convolutional neural network includes:

[0076] Obtain a sample set of bicycle shock absorbers, and divide the sample set into a training set and a test set;

[0077] Construct an attention mechanism module based on the Transformer module;

[0078] Construct a convolutional neural network, import the attention mechanism module into the convolutional neural network, and generate an improved convolutional neural network; in the early stage of the network, use a small convolution kernel (such as 3x3 or 5x5) and a small step size (such as a step size of 1). This can help capture detailed local features without losing spatial information prematurely, which is very important for accurately locating the focus point. As the depth of the network increases, the size of the convolution kernel can be gradually increased or the number of convolution layers can be increased to capture higher-level abstract features. These features can provide information about the overall structure and shape, which helps to identify specific parts of the shock absorber.

[0079] This embodiment uses maximum pooling to reduce the size of each feature map while retaining the most significant features, which is very helpful for highlighting important features (such as the focus area). However, the location and frequency of the pooling layer should be carefully selected to avoid excessive pooling that causes loss of important spatial information. Therefore, this embodiment uses an adaptive pooling layer at the end of the network, such as adaptive average pooling or adaptive maximum pooling, to ensure that the output feature map has a fixed size, which is particularly useful for subsequent fully connected layer processing and helps maintain performance consistency on input images of different sizes.

[0080] Training the improved convolutional neural network using the training set to obtain a training model;

[0081] Training the training model using the test set to generate training results;

[0082] Fine-tune the parameters of the training model based on the training results to generate the focus point activation analysis model;

[0083] The sample set of bicycle shock absorbers includes force point analysis information. In this embodiment, the collected sample set of bicycle shock absorbers is re-labeled with force points.

[0084] The process of marking the focus point includes:

[0085] Identify points of force: Identify which points are subject to the greatest forces, these are usually points of connection or fixing.

[0086] Decompose Force Vector: Decomposes the force vector into components perpendicular and parallel to the surface of interest.

[0087] Calculate stresses: Calculate the normal and shear stresses at each point using the formulas above.

[0088] Strain Calculation: Based on material properties such as Young's modulus and shear modulus, the strains caused by these stresses are calculated.

[0089] For a further optimized solution, the process of inputting the preprocessed image dataset into the force application point activation analysis model for calculation to obtain the high-activation region vector of the shock absorber force includes:

[0090] Extract the local force characteristics of the preprocessed image dataset through the attention mechanism module of the force application point activation analysis model to obtain the force attention score;

[0091] Obtain the force application point distribution vector based on the force attention score and the convolutional module of the force application point activation analysis model;

[0092] Calculate the correlation degree of the force application point distribution vector, and perform regional division on the force application point distribution vector based on the correlation degree to obtain the high-activation region vector of the shock absorber force.

[0093] For a further optimized solution, the expression for obtaining the force attention score is:

[0094]

[0095] where Score(f x ,f y ) represents the force attention score, f x represents the component in the working direction of the shock absorber, f y represents the component in the vertical direction of the shock absorber, T represents matrix transpose, represents the square root of the dimension of the component in the vertical direction of the shock absorber.

[0096] The Transformer structure calculates the force attention score. Based on the Transformer module, the global features of the bicycle shock absorber are obtained. Its input is the feature map with reduced size through the average pooling layer to reduce the computational cost. The specific process is that each pixel in the feature map is mapped to a high-dimensional embedding vector. Each pixel is regarded as a token and its features are represented by the embedding vector. Then, the token embedding operation is performed. Since the Transformer algorithm itself does not contain the position information of the sequence, additional positional encodings need to be added to represent the position information of each pixel. Among them, the positional encoding is a fixed vector, and its value changes according to the position of the pixel. By combining the positional encoding with the token embedding, it helps the model understand the different position information in the sequence.

[0097] The Transformer structure calculates the dot product of the query vector and the key vector to obtain the attention score. This embodiment replaces the query vector and the key vector with the components in the working direction of the shock absorber and the components in the vertical direction of the shock absorber to better scale the attention score, and applies the softmax function to obtain the attention weight. The self-attention mechanism is applied to the global information of the sequence to capture long-distance dependencies. Then, a convolutional neural network can be used to process local patterns and feature extraction. These two structures can be connected in series or in parallel.

[0098] It is also possible to consider adding residual connections to the network, which helps train deeper network structures without causing gradient vanishing or exploding. Residual connections also help maintain the integrity of input information in deeper layers, which is very important for maintaining accurate prediction of the focus point location.

[0099] The formula for calculating the grid position of the convolution kernel along the x-axis is as follows:

[0100]

[0101] The formula for calculating the grid position of the convolution kernel along the y-axis is as follows:

[0102]

[0103] Since the Δ offset is in decimal form and the coordinates are in integer form, the bilinear interpolation method is used:

[0104] CK=∑ CK′ B(CK′,CK)·CK′;

[0105] Where CK represents CK in the above formula i±c With CK j±c The decimal part, CK′ represents the enumeration of all integer spaces, and B represents the two-dimensional bilinear interpolation kernel, which is decomposed into two one-dimensional kernels. The formula is as follows:

[0106] B(CK,CK′)=b(CK x ,CK x ′)·b(CK y ,CK y ′);

[0107] The feature map size obtained by bilinear interpolation sampling along the x-axis is [B, C, K*H, W]. After the convolution kernel operation with a size of K×1 and the ReLU activation function, the output feature map size in the x-axis direction is [B, C, H, W].

[0108] Finally, extract the vectors of the central force points of the feature map, calculate the correlation of the force point distribution vectors, and perform regional division on the force point distribution vectors based on the correlation to obtain the high-activation region vectors of the shock absorber force.

[0109] For a further optimized solution, the process of constructing a 3D model of a bicycle shock absorber based on the high-activation region vectors of the shock absorber force and shock absorber parameters includes:

[0110] Construct a shock absorber model based on the shock absorber parameters;

[0111] Import the high-activation region vectors of the shock absorber force as the force attributes of the shock absorber into the shock absorber model to generate a shock absorber force model, and verify the structural strength and function of the shock absorber force model through the finite element analysis method to obtain the 3D model of the bicycle shock absorber.

[0112] Convolution module: Perform a convolution operation on the input data through a set of convolution kernels to extract spatial features. Parameters such as the size, stride, and padding of the convolution kernels can affect the scale of feature extraction.

[0113] Force point distribution vector: This vector can represent the "attention" or "importance" of different regions in the image. It can be obtained through the following steps:

[0114] Feature map generation: The feature map output by the convolution layer contains local features of the image at different positions.

[0115] Weighted aggregation: Generate a vector by weighting or aggregating features in different regions, and this vector represents the "force point distribution" of the entire image.

[0116] Fusion attention score: If an attention mechanism is used, the obtained attention score can be combined with the convolution feature map to enhance the model's attention to key regions.

[0117] At this time, the 3D model of the shock absorber is no longer a simple 3D model, but can simulate the stress and strain of the shock absorber at each point through the high-activation region vectors of the shock absorber force; use finite element analysis in CAD or specialized simulation software to verify the structural strength and function of the shock absorber. And it can simulate actual working conditions such as collisions and long-term fatigue to test the performance of the shock absorber.

[0118] Embodiment 2

[0119] As Figure 2 shown, in this embodiment, a modeling system for a bicycle shock absorber is provided, including:

[0120] A data acquisition subsystem for acquiring shock absorber images of different types and different working states of the bicycle shock absorber;

[0121] A data processing subsystem for annotating key parts of the shock absorber image, processing the annotated shock absorber image to obtain a preprocessed image dataset;

[0122] A calculation subsystem for constructing a force application point activation analysis model based on a convolutional neural network, inputting the preprocessed image dataset into the force application point activation analysis model for calculation to obtain a shock absorber force high activation region vector;

[0123] A modeling subsystem for constructing a three-dimensional model of a bicycle shock absorber based on the shock absorber force high activation region vector and shock absorber parameters.

[0124] The above is only a preferred specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed in the present application should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A modeling method for a bicycle shock absorber, characterized in that, Including the following steps: Obtain shock absorber images of different types and different working states of a bicycle shock absorber; Label the key parts of the shock absorber images, process the labeled shock absorber images, and obtain a preprocessed image dataset; Construct a force point activation analysis model based on a convolutional neural network, input the preprocessed image dataset into the force point activation analysis model for calculation, and obtain a high activation region vector of the shock absorber force; Construct a 3D model of the bicycle shock absorber based on the high activation region vector of the shock absorber force and shock absorber parameters; The process of constructing the force point activation analysis model based on the convolutional neural network includes: Obtain a sample set of bicycle shock absorbers, and divide the sample set into a training set and a test set; Construct an attention mechanism module based on the Transformer module; Construct a convolutional neural network, import the attention mechanism module into the convolutional neural network, and generate an improved convolutional neural network; Train the improved convolutional neural network with the training set to obtain a training model; Train the training model with the test set to generate a training result; Fine-tune the parameters of the training model based on the training result to generate the force point activation analysis model; The sample set of the bicycle shock absorber includes force point analysis information; The process of inputting the preprocessed image dataset into the force point activation analysis model for calculation and obtaining a high activation region vector of the shock absorber force includes: Extract the local force characteristics of the preprocessed image dataset through the attention mechanism module of the force point activation analysis model to obtain force attention scores; Obtain a force point distribution vector based on the force attention scores and the convolutional module of the force point activation analysis model; Calculate the correlation of the force point distribution vector, and divide the force point distribution vector based on the correlation to obtain a high activation region vector of the shock absorber force.

2. The modeling method of the bicycle shock absorber according to claim 1, characterized in that, The process of obtaining shock absorber images of different types and different working states of a bicycle shock absorber includes: Use a depth camera to capture different working states of different types of bicycle shock absorbers to obtain captured images; The different types of bicycle shock absorbers include, but are not limited to, spring shock absorbers and gas shock absorbers; The different working states include, but are not limited to, the compressed state and the uncompressed state.

3. The modeling method of the bicycle shock absorber according to claim 1, characterized in that, The process of labeling the key parts of the shock absorber images includes: Obtain the coordinates of the force points of the shock absorber in the shock absorber image to generate shock absorber force point coordinates; Convert the shock absorber force point coordinates into a force point vector, and decompose the force point vector into components along the working direction and the vertical direction of the shock absorber; Label the components along the working direction and the vertical direction of the shock absorber in the shock absorber image to complete the labeling of the key parts of the shock absorber image.

4. The modeling method of the bicycle shock absorber according to claim 1, characterized in that, The process of processing the labeled shock absorber images to obtain a preprocessed image dataset includes: Obtain the morphological characteristics of the shock absorber, and generate a pixel point threshold range based on the morphological characteristics of the shock absorber; Eliminate the pixel points that do not meet the pixel point threshold range in each pixel point of the marked shock absorber image to generate a shock absorber feature image; Unify the sizes of all shock absorber feature images to generate the preprocessed image dataset.

5. The modeling method of the bicycle shock absorber according to claim 1, characterized in that, The process of constructing a three-dimensional model of a bicycle shock absorber based on the high-activation region vector of the shock absorber force and shock absorber parameters includes: Construct a shock absorber model based on the shock absorber parameters; Import the high-activation region vector of the shock absorber force as the shock absorber force attribute into the shock absorber model to generate a shock absorber force model; Verify the structural strength and function of the shock absorber force model through the finite element analysis method to obtain the three-dimensional model of the bicycle shock absorber.

6. The modeling method of the bicycle shock absorber according to claim 1, wherein The expression for obtaining the force attention score is: ； Among them, represents the force attention score, represents the component of the shock absorber working direction, represents the component of the shock absorber in the vertical direction, T represents matrix transpose, represents the square root of the dimension of the component of the shock absorber in the vertical direction, x represents the shock absorber working direction, and y represents the shock absorber.

7. A modeling system for a bicycle shock absorber, which is used to execute the modeling method of the bicycle shock absorber according to any one of claims 1-6, characterized in that, Including: A data acquisition subsystem for acquiring shock absorber images of different types and different working states of bicycle shock absorbers; A data processing subsystem for annotating the key parts of the shock absorber image, processing the marked shock absorber image, and obtaining a preprocessed image dataset; A calculation subsystem for constructing a force point activation analysis model based on a convolutional neural network, inputting the preprocessed image dataset into the force point activation analysis model for calculation, and obtaining a high-activation region vector of the shock absorber force; A modeling subsystem for constructing a three-dimensional model of a bicycle shock absorber based on the high-activation region vector of the shock absorber force and shock absorber parameters.

Images