Acquisition method, device and equipment of along-line engineering control points and medium

Abstract

The invention provides a method, a device and equipment for acquiring control points of engineering along a line and a medium, and relates to the technical field of railway engineering. Comprising the following steps: carrying out image processing on image features of a parallel branch to obtain a feature map of the parallel branch; performing feature extraction and fusion on the feature maps of the parallel branches based on a decoder to obtain multi-scale features of the image; performing feature extraction on the image features of the parallel branches and the multi-scale features of the images according to a preset network deep supervision method to obtain a along-line ground feature model; based on a preset loss function and a preset network deep supervision method, carrying out optimization solution on the along-line ground feature model to obtain an optimal along-line ground feature model; and solving the optimal along-line ground object model according to a preset overlay analysis method to obtain a control point of the along-line project. The problems that a designer easily omits some important control elements in the reconnaissance data processing process, the control point data extraction efficiency is low and the like are solved.

Description

Technical Field

[0001] The present invention relates to the field of railway engineering technology, and in particular to a method, device, equipment and medium for acquiring engineering control points along a line. Background Art

[0002] In the existing railway engineering technology, survey and design personnel conduct surveys on roads, existing railways, rivers and buildings. Due to factors such as long survey lines, complex or changeable terrain and objects, designers are prone to miss some important control elements in the process of survey data processing, and cannot guarantee the problem that data can be extracted quickly and accurately. Therefore, there is an urgent need for a method, device, equipment and medium for obtaining engineering control points along the line, which solves the problem that factors such as long survey lines, complex or changeable terrain and objects cause designers to easily miss some important control elements in the process of survey data processing, and cannot guarantee the problem that data can be extracted quickly and accurately. Summary of the invention

[0003] The purpose of the present invention is to provide a method, device, equipment and medium for obtaining engineering control points along the line to improve the above problems. In order to achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0004] In the first aspect, the present application provides a method for acquiring engineering control points along the route, comprising:

[0005] Get the visual state space model;

[0006] Performing image extraction on preset multi-dimensional features based on the visual state space model to obtain image features of parallel branches;

[0007] Performing image processing on the image features of the parallel branches to obtain a feature map of the parallel branches;

[0008] Extracting and fusing the feature graphs of the parallel branches based on a preset decoder to obtain multi-scale features of the image;

[0009] Extracting features of the image of the parallel branch and multi-scale features of the image according to a preset network deep supervision method to obtain a model of terrain features along the line;

[0010] Based on a preset loss function and a preset network deep supervision method, the terrain model along the line is optimized and solved to obtain an optimal terrain model along the line;

[0011] The optimal terrain model along the line is solved according to a preset overlay analysis method to obtain control points of the project along the line.

[0012] In a second aspect, the present application also provides a device for acquiring engineering control points along the route, comprising:

[0013] A first acquisition module is used to acquire a visual state space model;

[0014] A first processing module, configured to extract images of preset multi-dimensional features based on the visual state space model to obtain image features of parallel branches;

[0015] A second processing module, used for performing image processing on the image features of the parallel branches to obtain a feature map of the parallel branches;

[0016] A third processing module is used to extract and fuse the feature graphs of the parallel branches based on a preset decoder to obtain multi-scale features of the image;

[0017] A fourth processing module is used to extract the image features of the parallel branch and the multi-scale features of the image according to a preset network deep supervision method to obtain a terrain model along the line;

[0018] A fifth processing module is used to optimize and solve the terrain model along the line based on a preset loss function and a preset network deep supervision method to obtain an optimal terrain model along the line;

[0019] The sixth processing module is used to solve the optimal terrain model along the line according to a preset overlay analysis method to obtain control points of the engineering along the line.

[0020] In a third aspect, the present application also provides a device for acquiring engineering control points along the route, including:

[0021] Memory for storing computer programs;

[0022] A processor is used to implement the steps of the method for acquiring engineering control points along the route when executing the computer program.

[0023] In a fourth aspect, the present application further provides a readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the above-mentioned method for acquiring control points along the engineering line are implemented.

[0024] The beneficial effects of the present invention are:

[0025] The present invention obtains the model of the objects along the line through the image features of the parallel branches and the multi-scale features of the images, optimizes and solves the model of the objects along the line according to the loss function and the network deep supervision method, and then processes it through the overlay analysis method to obtain the control points of the engineering along the line. It solves the problem that factors such as long survey lines and complex or changeable terrain objects cause designers to easily miss some important control elements during the survey data processing process, and cannot ensure that data can be extracted quickly and accurately.

[0026] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by implementing the embodiments of the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0028] Figure 1 It is a schematic flow chart of a method for acquiring engineering control points along the line described in an embodiment of the present invention;

[0029] Figure 2 It is a schematic diagram of the structure of the equipment for acquiring the engineering control points along the line described in an embodiment of the present invention.

[0030] Markings in the figure: 800, acquisition equipment of engineering control points along the line; 801, processor; 802, memory; 803, multimedia component; 804, I / O interface; 805, communication component. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0032] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0033] Embodiment 1:

[0034] This embodiment provides a method for acquiring engineering control points along a route.

[0035] See also Figure 1 , the figure shows that the method includes steps S1 to S7, including:

[0036] S1: Get the visual state space model;

[0037] S2: extracting images from preset multi-dimensional features based on the visual state space model to obtain image features of parallel branches;

[0038] The visual state space model includes a feature selection module.

[0039] In order to clarify the specific method of obtaining the image features of the parallel branches, step S2 includes S21 to S25, which are specifically:

[0040] S21: Acquire images along the project line;

[0041] S22: performing geometric correction and radiation correction on the images along the project to obtain a target image;

[0042] In this step, the geometric correction is used to eliminate or correct the geometric errors of remote sensing images, and the geometric correction includes coarse geometric correction, fine geometric correction, mixed correction and orthorectification; the radiation correction is the correction of systematic and random radiation distortion or distortion generated by the data acquisition and transmission system, and the process of eliminating or correcting image brightness distortion caused by radiation error.

[0043] S23: performing convolution and pooling operations on the target image based on a preset ground feature extraction network model to obtain initial features of the image;

[0044] In this step, the target image first undergoes a 3×3 convolution with a convolution kernel number of 32 and a step size of 2; the target image then undergoes a 3×3 convolution with a convolution kernel number of 32 and a step size of 1; the target image finally undergoes a 3×3 convolution with a convolution kernel number of 96 and a step size of 1 and a 3×3 maximum pooling with a step size of 2 to obtain the initial features of the image;

[0045] The initial features of the image are:

[0046]

[0047] In the above formula (1), H represents the height of the target image, W represents the width of the target image, and 96 represents the number of convolution kernels;

[0048] Preferably, the object extraction network model is ResNet Stem, and the Stem in the ResNet Stem is used to reduce the size of the input image, extract the preliminary features of the input image, reduce the amount of calculation of subsequent layers, improve network efficiency and stabilize the learning process.

[0049] S24: performing multi-dimensional feature extraction on the initial features of the image to obtain multi-dimensional features;

[0050] The ground feature extraction network model includes a deep residual network model;

[0051] In order to clarify the specific method of obtaining the multi-dimensional features, step S24 includes S241 to S247, which are:

[0052] S241: solving the initial features of the image according to the deep residual network model to obtain an initial feature map;

[0053] In this step, the initial features of the image are input into the deep residual network model, and the deep residual network model after the input is solved through the backbone part of ResNet in the ResNet Stem to obtain an initial feature map.

[0054] S242: Performing deformable convolution processing on the initial feature map to obtain a first branch feature;

[0055] S243: Processing the first branch feature based on the feature selection module to obtain a processed first branch feature;

[0056] Preferably, the feature selection modules are two consecutive VSS modules.

[0057] S244: performing deformable convolution processing on the processed first branch feature to obtain a second branch feature;

[0058] S245: performing integrated deep feature processing on the second branch feature to obtain an integrated second branch feature;

[0059] In this step, the integrated deep feature is upsampling and fusion of the first branch feature, the second branch feature is upsampled and fused with the first branch feature to obtain a processed second branch feature, and the feature selection module processes the processed second branch feature to obtain an integrated second branch feature;

[0060] Preferably, the feature selection modules are two consecutive VSS modules.

[0061] S246: performing a deformable convolution process on the first branch feature and the integrated second branch feature to obtain a third branch feature;

[0062] In this step, the fourth branch feature is also included, and the first branch feature, the second branch feature and the third branch feature are processed based on the feature selection module to obtain the fourth branch feature;

[0063] Preferably, the feature selection module is nine consecutive VSS modules.

[0064] S247: Taking the first branch feature, the second branch feature and the third branch feature as multi-dimensional features.

[0065] S25: Performing image extraction on the multi-dimensional features based on the feature selection module to obtain image features of the parallel branches.

[0066] Preferably, the feature selection modules are two consecutive VSS modules.

[0067] S3: performing image processing on the image features of the parallel branches to obtain a feature map of the parallel branches;

[0068] S4: extracting and fusing the feature graphs of the parallel branches based on a preset decoder to obtain multi-scale features of the image;

[0069] The decoder includes an attention module;

[0070] In order to clarify the specific method of obtaining the multi-scale features of the image, step S4 includes S41 to S45, which are specifically:

[0071] S41: extracting and fusing the feature graphs of the parallel branches according to the separation ratio preset in the attention module to obtain spatial features of the feature graphs and channel features of the feature graphs;

[0072] In this step, the separation ratio is two sets of feature maps of the separation ratio division channel, and the two sets of feature maps of the separation ratio division channel perform feature extraction and fusion on the feature maps of the parallel branches to obtain the spatial features of the feature maps and the channel features of the feature maps.

[0073] S42: performing convolution adjustment on the preset number of channels of the feature map of the parallel branch to obtain the number of convolution channels;

[0074] In this step, the convolution adjustment is 1×1 convolution adjustment.

[0075] S43: solving the feature graph of the parallel branch according to a preset logic function to obtain an attention feature of the feature graph;

[0076] The logic function includes a normalization function and an S-type growth curve function;

[0077] In order to clarify the specific method of obtaining the attention feature of the feature map, step S43 includes S431 to S436, which are:

[0078] S431: performing global average pooling processing on the feature map of the parallel branch to obtain an average pooling feature;

[0079] S432: performing global maximum pooling processing on the feature map of the parallel branch to obtain maximum pooling features;

[0080] S433: constructing based on the average pooling feature and the maximum pooling feature to obtain a mixed pooling feature;

[0081] Preferably, the average pooling feature and the maximum pooling feature are added to obtain a mixed pooling feature.

[0082] S434: performing mixed pooling feature processing on the feature map of the parallel branch according to the normalization function to obtain a channel attention size;

[0083] Preferably, the channel attention size is a weight in the range of 0-1.

[0084] S435: performing global average pooling processing on the feature map of the parallel branch according to the S-shaped growth curve function to obtain a spatial attention weight;

[0085] In this step, the spatial attention performs global average pooling on the feature map of the parallel branch in the x-direction and the y-direction respectively, and then processes it through the S-shaped growth curve function to obtain the spatial attention weight.

[0086] S436: Construct based on the channel attention size and the spatial attention weight to obtain the attention feature of the feature map.

[0087] S44: performing calculations on the feature graph of the parallel branch using different convolution kernels and different expansion rates based on the attention module to obtain a convolution data set;

[0088] In order to clarify the specific method of obtaining the convolution data set, step S44 includes S441 to S444, which are specifically:

[0089] S441: Input the feature map of the parallel branch into the attention module, perform upsampling processing on the feature map of the parallel branch by the bilinear interpolation method, and obtain the operation result of the first branch and the operation result of the second branch;

[0090] S442: performing calculations with different convolution kernels and different expansion rates on the operation result of the first branch to obtain first convolution data;

[0091] Preferably, the different convolution kernels and different expansion rates are calculated as convolution branches with different convolution kernels and different expansion rates; the convolution branches with different convolution kernels and different expansion rates include: convolution kernel 5 and expansion rate 1; convolution kernel 7 and expansion rate 1; convolution kernel 3 and expansion rate 6; convolution kernel 3 and expansion rate 12; convolution kernel 3 and expansion rate 18.

[0092] S443: performing calculations with different convolution kernels and different expansion rates on the operation result of the second branch to obtain second convolution data;

[0093] S444: Construct according to the first convolution data and the second convolution data to obtain a convolution data set.

[0094] In this step, all different convolution kernels and different expansion rates are added together, and then the first convolution data and the second convolution data are calculated to obtain a convolution data set.

[0095] S45: Constructing based on the spatial features of the feature map, the channel features of the feature map, the number of convolution channels, the attention features of the feature map and the convolution data set to obtain multi-scale features of the image.

[0096] In this step, construction is performed by upsampling.

[0097] S5: extracting features of the image features of the parallel branch and the multi-scale features of the image according to a preset network deep supervision method to obtain a model of terrain features along the line;

[0098] S6: Optimizing and solving the terrain model along the line based on a preset loss function and a preset network deep supervision method to obtain an optimal terrain model along the line;

[0099] The loss function includes a marginal loss function and a cross entropy loss function;

[0100] In order to clarify the specific method of obtaining the optimal terrain model along the line, step S6 includes S61 to S65, which are specifically:

[0101] S61: Acquire label contour information of a target image, wherein the label contour information of the target image includes a true label contour value, a first internal pixel number, and a second internal pixel number;

[0102] S62: Calculate the true label contour value based on a preset loss function to obtain a loss value;

[0103] In order to clarify the specific method of obtaining the loss value, step S62 includes S621 to S623, which are:

[0104] S621: Calculate based on the first number of internal pixels and the second number of internal pixels to obtain a total number of pixels;

[0105] In this step, the total number of pixels is:

[0106] N=|X|+|Y| (2)

[0107] In the above formula (2): N represents the total number of pixels, |X| represents the number of pixels inside X, and |Y| represents the number of pixels inside Y;

[0108] The X internal pixel number represents a first internal pixel number, and the Y internal pixel number represents a second internal pixel number.

[0109] S622: Calculate the true label contour value and the predicted label contour value based on the edge loss function to obtain an edge loss value of the sample pixel;

[0110] In this step, the edge loss value of the sample pixel is:

[0111]

[0112] In the above formula (3): L edge represents the edge loss value of the sample pixel, N represents the total number of pixels, x i Represents the predicted label contour value, y i Represents the true label contour value.

[0113] S623: Calculate the total number of pixels and the predicted pixel probability based on the cross entropy loss function to obtain a cross entropy loss value between the output result and the true label.

[0114] In this step, the cross entropy loss value between the output result and the true label is:

[0115]

[0116] In the above formula (4), CELoss represents the cross entropy loss value between the output result and the true label, N represents the total number of pixels, M represents the number of categories, is a one-hot label variable. Represents the probability that the predicted pixel i belongs to category c;

[0117] The predicted probability that pixel i belongs to category c represents the predicted pixel probability.

[0118] S63: Calculate the first number of internal pixels and the second number of internal pixels based on a preset network deep supervision method to obtain a sample result;

[0119] In this step, the sample results are:

[0120]

[0121] In the above formula (5): Loss dice (X,Y) represents the sample result of (X,Y) under deep supervision of the network; |X∩Y| represents the number of pixels in the intersection of X and Y, |X| represents the number of pixels inside X, |Y| represents the number of pixels inside Y, X represents the result of sample prediction, and Y represents the true label of the dataset;

[0122] The sample result under the network deep supervision represents the sample result, the X internal pixel number represents the first internal pixel number, and the Y internal pixel number represents the second internal pixel number.

[0123] S64: constructing a ground feature loss calculation formula according to the loss value, the sample result, the preset weight parameter and the regularization term of the preset weight parameter;

[0124] In this step, the ground object loss calculation formula is:

[0125]

[0126] In the above formula (6), Loss represents the ground object loss function, α1, α2 and α3 all represent learnable weight parameters, and Loss dice represents the sample result under deep network supervision, CELoss represents the cross entropy loss value between the output result and the true label, L edge Represents the edge loss value of the sample pixel, Represents the L2 regularization term calculated by the weight parameter;

[0127] The learnable weight parameter represents the weight parameter, and the L2 regularization term calculated by the weight parameter represents the regularization term of the preset weight parameter.

[0128] S65: Optimizing and solving the terrain model along the line based on the terrain loss calculation formula to obtain the optimal terrain model along the line.

[0129] S7: Solving the optimal terrain model along the line according to a preset overlay analysis method to obtain control points of the engineering along the line.

[0130] In order to clarify the specific method of obtaining the control points of the projects along the route, step S7 includes S71 to S74, which are:

[0131] S71: predicting remote sensing images for the optimal terrain model along the route to obtain a first terrain category and a second terrain category;

[0132] S72: constructing according to the first ground feature category and the second ground feature category to obtain a ground feature classification result;

[0133] In step, the first ground feature category and the second ground feature category are combined to obtain a ground feature classification result.

[0134] S73: Solving the ground object classification result based on a preset grid space data conversion library and a preset image processing library to obtain the ground object boundary;

[0135] In this step, the object classification results are converted into vector data, the threshold is set based on the raster spatial data conversion library to delete the fragmented parts of the vector data, and the holes in the vector data are removed based on the image processing library to obtain the object boundary.

[0136] S74: Solve the boundary of the land feature and the preset railway data based on the preset overlay analysis method to obtain the control points of the project along the line.

[0137] In this step, vector overlay analysis is performed on the vector data and the railway data to obtain control points of the project along the line.

[0138] Embodiment 2:

[0139] This embodiment provides a device for acquiring engineering control points along the route, the device comprising:

[0140] A first acquisition module is used to acquire a visual state space model;

[0141] A first processing module, configured to extract images of preset multi-dimensional features based on the visual state space model to obtain image features of parallel branches;

[0142] In order to clarify the specific methods of obtaining the image features of the parallel branches, the specific methods are:

[0143] The first acquisition unit is used to acquire images along the project line;

[0144] A first correction unit is used to perform geometric correction and radiation correction on the image along the project to obtain a target image;

[0145] A first processing unit is used to perform convolution and pooling operations on the target image based on a preset ground feature extraction network model to obtain initial features of the image;

[0146] A second processing unit is used to perform multi-dimensional feature extraction on the initial features of the image to obtain multi-dimensional features;

[0147] The third processing unit is used to perform image extraction on the multi-dimensional features based on the feature selection module to obtain image features of the parallel branches.

[0148] A second processing module, used for performing image processing on the image features of the parallel branches to obtain a feature map of the parallel branches;

[0149] A third processing module is used to extract and fuse the feature graphs of the parallel branches based on a preset decoder to obtain multi-scale features of the image;

[0150] A fourth processing module is used to extract the image features of the parallel branch and the multi-scale features of the image according to a preset network deep supervision method to obtain a terrain model along the line;

[0151] A fifth processing module is used to optimize and solve the terrain model along the line based on a preset loss function and a preset network deep supervision method to obtain an optimal terrain model along the line;

[0152] The sixth processing module is used to solve the optimal terrain model along the line according to a preset overlay analysis method to obtain control points of the engineering along the line.

[0153] In order to clarify the specific methods of obtaining the image features of the parallel branches, the specific methods are:

[0154] A first construction unit is used to predict the remote sensing image of the optimal terrain model along the line to obtain a first terrain category and a second terrain category;

[0155] A second construction unit is used to construct according to the first ground feature category and the second ground feature category to obtain a ground feature classification result;

[0156] A third construction unit is used to solve the ground object classification result based on a preset raster space data conversion library and a preset image processing library to obtain the ground object boundary;

[0157] The fourth construction unit is used to solve the boundary of the land feature and the preset railway data based on a preset overlay analysis method to obtain control points of the project along the line.

[0158] It should be noted that, regarding the device in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment of the method, and will not be elaborated here.

[0159] Embodiment 3:

[0160] Corresponding to the above method embodiment, this embodiment also provides a device for acquiring engineering control points along the line. The device for acquiring engineering control points along the line described below and the method for acquiring engineering control points along the line described above can refer to each other.

[0161] Figure 2FIG. 8 is a block diagram of a device 800 for acquiring engineering control points along a route according to an exemplary embodiment. Figure 2 As shown, the device 800 for acquiring engineering control points along the line may include: a processor 801 and a memory 802. The device 800 for acquiring engineering control points along the line may also include one or more of a multimedia component 803, an I / O interface 804, and a communication component 805.

[0162] The processor 801 is used to control the overall operation of the acquisition device 800 of the engineering control points along the line, so as to complete all or part of the steps in the acquisition method of the engineering control points along the line. The memory 802 is used to store various types of data to support the operation of the acquisition device 800 of the engineering control points along the line, and these data may include, for example, instructions for any application or method used to operate on the acquisition device 800 of the engineering control points along the line, and application-related data, such as contact data, sent and received messages, pictures, audio, video, etc. The memory 802 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk. The multimedia component 803 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may be further stored in the memory 802 or sent via the communication component 805. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 804 provides an interface between the processor 801 and other interface modules, and the above-mentioned other interface modules can be keyboards, mice, buttons, etc. These buttons can be virtual buttons or physical buttons. The communication component 805 is used for wired or wireless communication between the acquisition device 800 of the engineering control point along the line and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 805 can include: Wi-Fi module, Bluetooth module, NFC module.

[0163] In an exemplary embodiment, the device 800 for acquiring the engineering control points along the line can be implemented by one or more application specific integrated circuits (Application Specific Integrated Circuit, referred to as ASIC), digital signal processors (Digital Signal Processor, referred to as DSP), digital signal processing devices (Digital Signal Processing Device, referred to as DSPD), programmable logic devices (Programmable Logic Device, referred to as PLD), field programmable gate arrays (Field Programmable Gate Array, referred to as FPGA), controllers, microcontrollers, microprocessors or other electronic components, and is used to execute the above-mentioned method for acquiring the engineering control points along the line.

[0164] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned method for acquiring engineering control points along the line are implemented. For example, the computer-readable storage medium may be the above-mentioned memory 802 including program instructions, and the above-mentioned program instructions may be executed by the processor 801 of the acquisition device 800 for acquiring engineering control points along the line to complete the above-mentioned method for acquiring engineering control points along the line.

[0165] Embodiment 4:

[0166] Corresponding to the above method embodiment, a readable storage medium is also provided in this embodiment. The readable storage medium described below and the method for acquiring engineering control points along the line described above can refer to each other.

[0167] A readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the method for acquiring engineering control points along the line in the above method embodiment.

[0168] The readable storage medium may specifically be a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or other readable storage medium that can store program codes.

[0169] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0170] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for obtaining engineering control points along a line, characterized in that: include: Get the visual state space model; Performing image extraction on preset multi-dimensional features based on the visual state space model to obtain image features of parallel branches; Performing image processing on the image features of the parallel branches to obtain a feature map of the parallel branches; Extracting and fusing the feature graphs of the parallel branches based on a preset decoder to obtain multi-scale features of the image; Extracting features of the image of the parallel branch and multi-scale features of the image according to a preset network deep supervision method to obtain a model of terrain features along the line; Based on a preset loss function and a preset network deep supervision method, the terrain model along the line is optimized and solved to obtain an optimal terrain model along the line; The optimal terrain model along the line is solved according to a preset overlay analysis method to obtain control points of the project along the line.

2. The method for acquiring engineering control points along the line according to claim 1 is characterized in that the visual state space model includes a feature selection module, and image extraction is performed on preset multidimensional features based on the visual state space model to obtain image features of the parallel branch, including: Obtain images along the project route; Performing geometric correction and radiation correction on the images along the project to obtain the target image; Performing convolution and pooling operations on the target image based on a preset ground feature extraction network model to obtain initial features of the image; Performing multi-dimensional feature extraction on the initial features of the image to obtain multi-dimensional features; Based on the feature selection module, image extraction is performed on the multi-dimensional features to obtain image features of the parallel branches.

3. The method for acquiring engineering control points along the line according to claim 2, characterized in that the ground feature extraction network model includes a deep residual network model, and multi-dimensional feature extraction is performed on the initial features of the image to obtain multi-dimensional features, including: Solving the initial features of the image according to the deep residual network model to obtain an initial feature map; Performing deformable convolution processing on the initial feature map to obtain a first branch feature; Processing the first branch feature based on the feature selection module to obtain a processed first branch feature; Performing deformable convolution processing on the processed first branch feature to obtain a second branch feature; Performing integrated deep feature processing on the second branch feature to obtain an integrated second branch feature; Performing deformable convolution processing on the first branch feature and the integrated second branch feature to obtain a third branch feature; The first branch feature, the second branch feature and the third branch feature are used as multi-dimensional features.

4. The method for acquiring engineering control points along the line according to claim 2 is characterized in that the decoder includes an attention module, and based on a preset decoder, feature extraction and fusion are performed on the feature map of the parallel branch to obtain multi-scale features of the image, including: Extracting and fusing the feature graphs of the parallel branches according to the separation ratio preset in the attention module to obtain spatial features of the feature graphs and channel features of the feature graphs; Performing convolution adjustment on the preset number of channels of the feature graph of the parallel branch to obtain the number of convolution channels; Solving the feature graph of the parallel branch according to a preset logic function to obtain an attention feature of the feature graph; Based on the attention module, different convolution kernels and different expansion rates are calculated on the feature map of the parallel branch to obtain a convolution data set; Based on the spatial features of the feature map, the channel features of the feature map, the number of convolution channels, the attention features of the feature map and the convolution data set, a multi-scale feature of the image is obtained.

5. The method for obtaining control points of engineering along the line according to claim 1 is characterized in that the optimal terrain model along the line is solved according to a preset overlay analysis method to obtain the control points of the engineering along the line, including: Predicting remote sensing images of the optimal terrain model along the route to obtain a first terrain category and a second terrain category; Constructing according to the first ground feature category and the second ground feature category to obtain a ground feature classification result; Solving the ground object classification result based on a preset raster space data conversion library and a preset image processing library to obtain the ground object boundary; Based on a preset overlay analysis method, the boundary of the land feature and the preset railway data are solved to obtain control points of the project along the line.

6. A device for acquiring engineering control points along the line, characterized in that: include: A first acquisition module is used to acquire a visual state space model; A first processing module, configured to extract images of preset multi-dimensional features based on the visual state space model to obtain image features of parallel branches; A second processing module, used for performing image processing on the image features of the parallel branches to obtain a feature map of the parallel branches; A third processing module is used to extract and fuse the feature graphs of the parallel branches based on a preset decoder to obtain multi-scale features of the image; A fourth processing module is used to extract the image features of the parallel branch and the multi-scale features of the image according to a preset network deep supervision method to obtain a terrain model along the line; A fifth processing module is used to optimize and solve the terrain model along the line based on a preset loss function and a preset network deep supervision method to obtain an optimal terrain model along the line; The sixth processing module is used to solve the optimal terrain model along the line according to a preset overlay analysis method to obtain control points of the engineering along the line.

7. The device for acquiring engineering control points along the line according to claim 6, characterized in that the first processing module is used to extract the preset multi-dimensional features based on the visual state space model to obtain the image features of the parallel branch, including: The first acquisition unit is used to acquire images along the project line; A first correction unit is used to perform geometric correction and radiation correction on the image along the project to obtain a target image; A first processing unit is used to perform convolution and pooling operations on the target image based on a preset ground feature extraction network model to obtain initial features of the image; A second processing unit is used to perform multi-dimensional feature extraction on the initial features of the image to obtain multi-dimensional features; The third processing unit is used to perform image extraction on the multi-dimensional features based on the feature selection module to obtain image features of the parallel branches.

8. The device for acquiring control points of engineering along the line according to claim 6, characterized in that the sixth processing module is used to solve the optimal terrain model along the line according to a preset overlay analysis method to obtain the control points of the engineering along the line, including: A first construction unit is used to predict the remote sensing image of the optimal terrain model along the line to obtain a first terrain category and a second terrain category; A second construction unit is used to construct according to the first ground feature category and the second ground feature category to obtain a ground feature classification result; A third construction unit is used to solve the ground object classification result based on a preset raster space data conversion library and a preset image processing library to obtain the ground object boundary; The fourth construction unit is used to solve the boundary of the land feature and the preset railway data based on a preset overlay analysis method to obtain control points of the project along the line.

9. A device for acquiring engineering control points along the line, characterized in that: include: Memory for storing computer programs; A processor is used to implement the steps of the method for acquiring engineering control points along the line as described in any one of claims 1 to 5 when executing the computer program.

10. A readable storage medium, characterized in that : A computer program is stored on the readable storage medium, and when the computer program is executed by the processor, the steps of the method for obtaining engineering control points along the line as described in any one of claims 1 to 5 are implemented.

Images