Method for Controlling Hopper Inclination of Grouting Device Based on Set Difference of Shape Features

Abstract

The present application discloses a method for controlling the inclination of a barrel of a grouting device based on set difference of shape features, which includes: acquiring an image of the cross-section of a steel structure and the connection area between the steel structure and the ground; respectively passing the image of the cross-section of the steel structure and the image of the connection area through N convolutional layers to obtain a first convolutional feature map and a second convolutional feature map, where N is a positive integer greater than or equal to 2 and less than or equal to 3; calculating the difference between the first convolutional feature map and the second convolutional feature map to obtain a difference feature map; passing the difference feature map through a convolutional neural network to obtain a depth feature map; passing the depth feature map through a plurality of fully connected layers to obtain eigenvalues, and the output of the last fully connected layer in the plurality of fully connected layers is a single bit; and determining the target inclination angle of the barrel based on the eigenvalues.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence, and more specifically, to a method, system, and electronic device for controlling the inclination of a barrel of a grouting device based on set difference of shape features. Background Art

[0002] A steel structure is a structure composed of steel materials and is a type of building structure. During the assembly of existing steel structures, a grouting device is required for foundation pouring, and the grouting device injects slurry into the connection area between the steel structure and the ground. Such a grouting device usually includes a barrel, such as the slurry-containing barrel of a cement tanker, and the slurry is poured out of the barrel by tilting the barrel for grouting.

[0003] During the current construction process, the inclination of the barrel is judged by experience. Although the inclination of the barrel not reaching the optimal value will not cause too many problems, it still affects the construction process. For example, if the inclination angle is too small, the poured slurry is often insufficient, affecting the grouting speed, and if the inclination angle is too large, the poured slurry cannot evenly fill the connection area, affecting the pouring strength.

[0004] Therefore, an optimized technical solution for controlling the inclination angle of the barrel of the grouting device is expected.

[0005] Currently, deep learning and neural networks have been widely applied in fields such as computer vision, natural language processing, and text signal processing. In addition, deep learning and neural networks have also shown levels close to or even exceeding those of humans in fields such as image classification, object detection, semantic segmentation, and text translation.

[0006] In recent years, the development of deep learning and neural networks has provided new solutions and ideas for controlling the inclination angle of the barrel of the grouting device. Summary of the Invention

[0007] To solve the above technical problems, this application is proposed. Embodiments of this application provide a method, system, and electronic device for controlling the inclination of a barrel of a grouting device based on set difference of shape features, which determine the inclination angle of the barrel of the grouting device based on the set difference in the high-dimensional feature space of the shape features of the cross-section of the steel structure and the shape features of the connection area between the steel structure and the ground by a deep neural network.

[0008] According to one aspect of this application, a method for controlling the inclination of a barrel of a grouting device based on set difference of shape features is provided, which includes:

[0009] Obtaining images of the cross-section of the steel structure and the connection area between the steel structure and the ground, where the images of the cross-section of the steel structure and the connection area have the same scale;

[0010] The images of the cross-section of the steel structure and the image of the connection area are respectively passed through N convolutional layers to obtain a first convolutional feature map and a second convolutional feature map, where N is a positive integer greater than or equal to 2 and less than or equal to 3;

[0011] Calculate the difference between the first convolutional feature map and the second convolutional feature map to obtain a difference feature map, which represents the representation after the set difference operation of the shape features of the cross-section of the steel structure and the connection area in the high-dimensional feature space;

[0012] Pass the difference feature map through a convolutional neural network to obtain a depth feature map;

[0013] Pass the depth feature map through multiple fully connected layers to obtain an eigenvalue, where the output of the last fully connected layer in the multiple fully connected layers is one bit; and

[0014] Based on the eigenvalue, determine the target tilt angle of the barrel.

[0015] In the above-mentioned method for controlling the tilt of the barrel of the grouting device based on the set difference of shape features, when acquiring the image of the cross-section of the steel structure and the image of the connection area between the steel structure and the ground, the images of the cross-section of the steel structure and the image of the connection area are collected under the same shooting conditions.

[0016] In the above-mentioned method for controlling the tilt of the barrel of the grouting device based on the set difference of shape features, acquiring the image of the cross-section of the steel structure and the image of the connection area between the steel structure and the ground includes: acquiring the first depth of field when shooting the cross-section of the steel structure and the second depth of field when shooting the connection area; and, based on the first depth of field or the second depth of field, converting the images of the cross-section of the steel structure and the image of the connection area into images of the same scale.

[0017] In the above-mentioned method for controlling the tilt of the barrel of the grouting device based on the set difference of shape features, calculating the difference between the first convolutional feature map and the second convolutional feature map to obtain a difference feature map includes: subtracting the first convolutional feature map from the second convolutional feature map by pixel position to obtain the difference feature map.

[0018] In the above-mentioned method for controlling the tilt of the barrel of the grouting device based on the set difference of shape features, based on the eigenvalue, determining the target tilt angle of the barrel includes: normalizing the eigenvalue according to the value range of the eigenvalue obtained during the training process; and, mapping the normalized eigenvalue to the interval of the minimum and maximum values of the tilt angle of the barrel to obtain the target tilt angle of the barrel.

[0019] In the method for controlling the inclination of the barrel of the grouting device based on set difference of shape features, the method further includes: obtaining the current inclination angle of the barrel; and calculating the difference between the current inclination angle and the target inclination angle to obtain an inclination control angle.

[0020] According to another aspect of the present application, there is provided a system for controlling the inclination of the barrel of a grouting device based on set difference of shape features, including:

[0021] An image acquisition unit for acquiring images of the cross-section of the steel structure and the connection area between the steel structure and the ground, wherein the images of the cross-section of the steel structure and the connection area have the same scale;

[0022] A convolutional feature map generation unit for respectively passing the images of the cross-section of the steel structure and the connection area obtained by the image acquisition unit through N convolutional layers to obtain a first convolutional feature map and a second convolutional feature map, where N is a positive integer greater than or equal to 2 and less than or equal to 3;

[0023] A differential feature map generation unit for calculating the difference between the first convolutional feature map and the second convolutional feature map obtained by the convolutional feature map generation unit to obtain a differential feature map, where the differential feature map represents the representation after performing a set difference operation on the shape features of the cross-section of the steel structure and the connection area in a high-dimensional feature space;

[0024] A depth feature map generation unit for passing the differential feature map obtained by the differential feature map generation unit through a convolutional neural network to obtain a depth feature map;

[0025] An eigenvalue generation unit for passing the depth feature map obtained by the depth feature map generation unit through a plurality of fully connected layers to obtain eigenvalues, where the output of the last fully connected layer in the plurality of fully connected layers is one-bit; and

[0026] A target inclination angle generation unit for determining the target inclination angle of the barrel based on the eigenvalues obtained by the eigenvalue generation unit.

[0027] In the above system for controlling the inclination of the barrel of a grouting device based on set difference of shape features, the images of the cross-section of the steel structure and the connection area are acquired under the same shooting conditions.

[0028] In the cylinder tilting control system of the grouting device based on set difference of shape features described above, the image acquisition unit is further configured to: acquire a first depth of field when photographing the cross-section of the steel structure and a second depth of field when photographing the connection area; and convert the image of the cross-section of the steel structure and the image of the connection area into images of the same scale based on the first depth of field or the second depth of field.

[0029] In the cylinder tilting control system of the grouting device based on set difference of shape features described above, the differential feature map generation unit is further configured to: subtract the first convolutional feature map from the second convolutional feature map by pixel position to obtain the differential feature map.

[0030] In the cylinder tilting control system of the grouting device based on set difference of shape features described above, the target tilt angle generation unit includes: a normalization processing subunit configured to perform normalization processing on the eigenvalue according to the value range of the eigenvalue obtained during the training process; and a numerical mapping subunit configured to map the normalized eigenvalue into the range of the minimum and maximum values of the tilt angle of the cylinder to obtain the target tilt angle of the cylinder.

[0031] In the cylinder tilting control system of the grouting device based on set difference of shape features described above, the control system further includes: a tilt control unit configured to: acquire the current tilt angle of the cylinder; and calculate the difference between the current tilt angle and the target tilt angle to obtain a tilt control angle.

[0032] According to another aspect of the present application, there is provided an electronic device, including: a processor; and a memory in which computer program instructions are stored, and when the computer program instructions are run by the processor, the processor is caused to execute the cylinder tilting control method of the grouting device based on set difference of shape features as described above.

[0033] According to yet another aspect of the present application, there is provided a computer-readable medium having computer program instructions stored thereon, and when the computer program instructions are run by a processor, the processor is caused to execute the cylinder tilting control method of the grouting device based on set difference of shape features as described above.

[0034] According to the cylinder tilting control method, system and electronic device of the grouting device based on set difference of shape features provided by the present application, the tilt angle of the cylinder of the grouting device is determined by set difference of the shape features of the cross-section of the steel structure and the shape features of the connection area between the steel structure and the ground in a high-dimensional feature space by a deep neural network. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The above and other objects, features, and advantages of the present application will become more apparent by describing the embodiments of the present application in more detail with reference to the accompanying drawings. The drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation to the present application. In the drawings, the same reference numerals generally represent the same components or steps.

[0036] Figure 1 A schematic diagram of a scenario of a method for controlling the inclination of a barrel of a grouting device based on set difference of shape features according to an embodiment of the present application is illustrated.

[0037] Figure 2 A flowchart of a method for controlling the inclination of a barrel of a grouting device based on set difference of shape features according to an embodiment of the present application is illustrated.

[0038] Figure 3 A schematic diagram of an architecture of a method for controlling the inclination of a barrel of a grouting device based on set difference of shape features according to an embodiment of the present application is illustrated.

[0039] Figure 4 A flowchart of determining a target inclination angle of the barrel based on the eigenvalue in a method for controlling the inclination of a barrel of a grouting device based on set difference of shape features according to an embodiment of the present application is illustrated.

[0040] Figure 5 A block diagram of a control system for the inclination of a barrel of a grouting device based on set difference of shape features according to an embodiment of the present application is illustrated.

[0041] Figure 6 A block diagram of a target inclination angle generation unit in a control system for the inclination of a barrel of a grouting device based on set difference of shape features according to an embodiment of the present application is illustrated.

[0042] Figure 7 A block diagram of an electronic device according to an embodiment of the present application is illustrated. Detailed implementation manners

[0043] Next, exemplary embodiments according to the present application will be described in detail with reference to the drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. It should be understood that the present application is not limited by the exemplary embodiments described herein.

[0044] Scene Overview

[0045] As described above, a steel structure is a structure composed of steel materials and is a type of building structure. When assembling an existing steel structure, a grouting device is required for foundation pouring, and the slurry is injected into the connection area between the steel structure and the ground through the grouting device. Such a grouting device usually includes a material barrel, such as the slurry-containing barrel of a cement tanker, and the slurry is poured out of the barrel by tilting the barrel for grouting.

[0046] During the current construction process, the tilting of the barrel is judged by experience. Although the tilting of the barrel will not cause too many problems even if it is not optimal, it will still affect the construction process. For example, if the tilting angle is too small, the poured slurry is often insufficient, affecting the grouting speed, and if the tilting angle is too large, the poured slurry cannot evenly fill the connection area, affecting the pouring strength.

[0047] Therefore, an optimized technical solution for controlling the tilting angle of the barrel of the grouting device is expected.

[0048] The applicant of this application uses deep learning to learn the tilting angle of the barrel. First, the applicant of this application considers that the grouting process is related to the shape characteristics of the space into which the slurry is to be injected, and the shape characteristics of the space to be injected directly depend on the cross-sectional shape of the connection between the steel structure and the ground and the shape of the connection area on the ground. Therefore, it is considered to learn the relationship between the cross-sectional shape of the steel structure and the shape of the connection area on the ground and the tilting angle of the barrel through deep learning.

[0049] However, when the steel structure has been assembled, that is, when it has actually come into contact with the connection area on the ground, the obtained image has insufficient necessary key information due to shooting angle limitations, occlusion, etc., resulting in insufficient learning effectiveness. Therefore, the applicant of this application considers separately obtaining the cross-section of the steel structure and the image of the connection area between the steel structure and the ground, and then expressing the shape characteristics of the space into which the slurry is to be injected during the above grouting process through the difference operation of the feature distribution sets in their feature spaces.

[0050] Specifically, first obtain the cross-section of the steel structure and the image of the connection area between the steel structure and the ground, noting that the images need to be obtained under the same imaging conditions or need to be converted to the same imaging conditions, especially the distance from the lens to the imaging object, to ensure that the sizes of the two images are the same. Then, pass both of them through two to three convolutional layers to extract the feature maps for expressing their shape characteristics, and subtract the feature map of the cross-section of the steel structure from the feature map of the connection area on the ground. Here, since the shape characteristics of the space into which the slurry is to be injected here are shallow features, in order to improve the accuracy of coding, the obtained differential feature map is used to extract high-dimensional shape features through a convolutional neural network, and then through an encoder, that is, multiple fully connected layers, where the output of the last fully connected layer is one bit, to obtain the feature value.

[0051] Then, normalize the eigenvalue to the range of 0 to 1 according to its value range obtained during the training process, and then map it to the range from the minimum value to the maximum value of the inclination angle, so as to obtain the corresponding inclination angle size.

[0052] Based on this, the present application proposes a method for controlling the inclination of the barrel of a grouting device based on set difference of shape features, which includes: acquiring images of the cross-section of a steel structure and the connection area between the steel structure and the ground, wherein the images of the cross-section of the steel structure and the connection area have the same scale; respectively passing the images of the cross-section of the steel structure and the connection area through N convolutional layers to obtain a first convolutional feature map and a second convolutional feature map, where N is a positive integer greater than or equal to 2 and less than or equal to 3; calculating the difference between the first convolutional feature map and the second convolutional feature map to obtain a difference feature map, and the difference feature map represents the representation after the set difference operation of the shape features of the cross-section of the steel structure and the connection area in the high-dimensional feature space; passing the difference feature map through a convolutional neural network to obtain a depth feature map; passing the depth feature map through multiple fully connected layers to obtain an eigenvalue, where the output of the last fully connected layer in the multiple fully connected layers is a single bit; and, based on the eigenvalue, determining the target inclination angle of the barrel.

[0053] Figure 1 The figure shows a schematic scenario diagram of a method for controlling the inclination of the barrel of a grouting device based on set difference of shape features according to an embodiment of the present application.

[0054] As Figure 1 shown, in this application scenario, images of the cross-section of a steel structure (e.g., S as shown in Figure 1 ) are collected by a camera (e.g., C as shown in Figure 1 ), and images of the connection area between the steel structure and the ground (e.g., G as shown in Figure 1 ), wherein when collecting the image of the cross-section of the steel structure, the steel structure is attached to the ground; then, the image of the cross-section of the steel structure and the image of the connection area between the steel structure and the ground are input into a server (e.g., S1 as shown in Figure 1 ) deployed with an algorithm for controlling the inclination of the barrel of a grouting device based on set difference of shape features, where the server can process the image of the cross-section of the steel structure and the image of the connection area between the steel structure and the ground based on the algorithm for controlling the inclination of the barrel of a grouting device based on set difference of shape features to output the target inclination angle of the barrel. Furthermore, based on the target inclination angle of the barrel and the current inclination angle of the barrel, the inclination control angle of the barrel can be obtained.

[0055] After introducing the basic principle of the present application, various non-limiting embodiments of the present application will be specifically introduced with reference to the accompanying drawings.

[0056] Exemplary Method

[0057] Figure 2 The figure shows a flowchart of a method for controlling the inclination of a barrel of a grouting device based on set difference of shape features according to an embodiment of the present application. As Figure 2 shown, the method for controlling the inclination of a barrel of a grouting device based on set difference of shape features according to an embodiment of the present application includes: S110, obtaining images of the cross-section of a steel structure and the connection area between the steel structure and the ground, wherein the images of the cross-section of the steel structure and the connection area have the same scale; S120, respectively passing the images of the cross-section of the steel structure and the connection area through N convolutional layers to obtain a first convolutional feature map and a second convolutional feature map, wherein N is a positive integer greater than or equal to 2 and less than or equal to 3; S130, calculating the difference between the first convolutional feature map and the second convolutional feature map to obtain a difference feature map, and the difference feature map represents the representation after performing set difference operation on the shape features of the cross-section of the steel structure and the connection area in a high-dimensional feature space; S140, passing the difference feature map through a convolutional neural network to obtain a depth feature map; S150, passing the depth feature map through a plurality of fully connected layers to obtain a feature value, wherein the output of the last fully connected layer in the plurality of fully connected layers is one bit; and S160, based on the feature value, determining the target inclination angle of the barrel.

[0058] Figure 3 The figure shows a schematic structural diagram of a method for controlling the inclination of a barrel of a grouting device based on set difference of shape features according to an embodiment of the present application. As Figure 3 shown, in this network architecture, first, the obtained images of the cross-section of a steel structure and the connection area between the steel structure and the ground (for example, IN0 and IN1 as shown in Figure 3 ) are respectively passed through N convolutional layers (for example, CL as shown in Figure 3 ) to obtain a first convolutional feature map (for example, F1 as shown in Figure 3 ) and a second convolutional feature map (for example, F2 as shown in Figure 3 ), wherein N is a positive integer greater than or equal to 2 and less than or equal to 3; then, calculate the difference between the first convolutional feature map and the second convolutional feature map to obtain a difference feature map (for example, Fi as shown in Figure 3 ), and the difference feature map represents the representation after performing set difference operation on the shape features of the cross-section of the steel structure and the connection area in a high-dimensional feature space; then, pass the difference feature map through a convolutional neural network (for example, as shown inFigure 3 the CNN shown in [FIGURE] to obtain a depth feature map (e.g., Fd as shown in [FIGURE]); then, pass the depth feature map through a plurality of fully connected layers (e.g., Fcl as shown in [FIGURE]) to obtain a feature value, wherein the output of the last fully connected layer in the plurality of fully connected layers is one bit; finally, based on the feature value, determine the target tilt angle of the cartridge. Figure 3 the CNN shown in [FIGURE] to obtain a depth feature map (e.g., Fd as shown in [FIGURE]); then, pass the depth feature map through a plurality of fully connected layers (e.g., Fcl as shown in [FIGURE]) to obtain a feature value, wherein the output of the last fully connected layer in the plurality of fully connected layers is one bit; finally, based on the feature value, determine the target tilt angle of the cartridge. Figure 3 the CNN shown in [FIGURE] to obtain a depth feature map (e.g., Fd as shown in [FIGURE]); then, pass the depth feature map through a plurality of fully connected layers (e.g., Fcl as shown in [FIGURE]) to obtain a feature value, wherein the output of the last fully connected layer in the plurality of fully connected layers is one bit; finally, based on the feature value, determine the target tilt angle of the cartridge.

[0059] In step S110, acquire images of the cross-section of the steel structure and the connection area between the steel structure and the ground, wherein the images of the cross-section of the steel structure and the images of the connection area have the same scale. As described above, considering that the grouting process is related to the shape characteristics of the space where the slurry is to be injected, and the shape characteristics of the space to be injected directly depend on the cross-sectional shape of the connection between the steel structure and the ground and the shape of the connection area with the ground, therefore, collect the images of the cross-section of the steel structure and the images of the connection area between the steel structure and the ground as source domain data to determine the tilt control angle of the cartridge of the grouting device.

[0060] Specifically, in a specific implementation, when acquiring the images of the cross-section of the steel structure and the images of the connection area through a camera, the steel structure should be in contact with the ground. And, to ensure that the images of the cross-section of the steel structure and the images of the connection area have the same scale, the images of the cross-section of the steel structure and the images of the connection area are obtained under the same imaging conditions, especially the distance from the lens to the imaging object.

[0061] Of course, in other examples of the present application, the images of the cross-section of the steel structure and the images of the connection area may also be obtained under different imaging conditions. In this case, the acquired images need to be transformed so that the images of the cross-section of the steel structure and the images of the connection area have the same scale. For example, the image of the cross-section of the steel structure is taken at a first depth of field, and the image of the connection area is taken at a second depth of field. At this time, the images of the cross-section of the steel structure and the images of the connection area can be transformed into images of the same scale based on the first depth of field or the second depth of field. Specifically, when the first depth of field is greater than the second depth of field, the image of the cross-section of the steel structure can be upsampled or the image of the connection area can be downsampled so that the images of the cross-section of the steel structure and the images of the connection area have the same scale; when the first depth of field is less than the second depth of field, the image of the cross-section of the steel structure can be downsampled or the image of the connection area can be upsampled so that the images of the cross-section of the steel structure and the images of the connection area have the same scale.

[0062] That is, in some specific examples of the present application, the process of obtaining images of the cross-section of the steel structure and the connection area between the steel structure and the ground includes: obtaining a first depth of field when photographing the cross-section of the steel structure and a second depth of field when photographing the connection area; and converting the images of the cross-section of the steel structure and the image of the connection area into images of the same scale based on the first depth of field or the second depth of field.

[0063] In step S120, the images of the cross-section of the steel structure and the image of the connection area are respectively passed through N convolutional layers to obtain a first convolutional feature map and a second convolutional feature map, where N is a positive integer greater than or equal to 2 and less than or equal to 3. That is, the shape features in the images of the cross-section of the steel structure and the image of the connection area are extracted by 2 or 3 convolutional layers, that is, the first convolutional feature map and the second convolutional feature map.

[0064] Those skilled in the art should be aware that in terms of feature extraction by a convolutional neural network, the shape features are extracted in the first one to three layers. Therefore, in the technical solution of the present application, convolutional layers having the same structure as the first 2 to 3 layers of the convolutional neural network are used to process the images of the cross-section of the steel structure and the image of the connection area, so as to extract convolutional feature maps representing the shape features of the interface of the steel structure and the shape features of the connection area.

[0065] In step S130, the difference between the first convolutional feature map and the second convolutional feature map is calculated to obtain a difference feature map, and the difference feature map represents the representation after the set difference operation of the shape features of the cross-section of the steel structure and the connection area in the high-dimensional feature space. As mentioned above, when the steel structure has been assembled, that is, when it is actually in contact with the ground connection area, the obtained images have insufficient necessary key information due to shooting angle limitations, occlusion, etc., resulting in insufficient learning effectiveness. Therefore, the applicant of the present application considers separately obtaining images of the cross-section of the steel structure and the connection area of the ground, and then expressing the shape features of the space to be injected with slurry during the above grouting process through the difference operation of the feature distributions of their sets in the feature space.

[0066] Specifically, in the embodiment of the present application, the process of calculating the difference between the first convolutional feature map and the second convolutional feature map to obtain a difference feature map includes: subtracting the feature value at the corresponding position in the first convolutional feature map from the feature value of the second convolutional feature map at the pixel position to obtain the difference feature map. The difference feature map represents the representation after the set difference operation of the shape features of the cross-section of the steel structure and the connection area in the high-dimensional feature space. Corresponding to this application scenario, the difference feature map represents the shape features of the space to be injected.

[0067] In step S140, the differential feature map is passed through a convolutional neural network to obtain a depth feature map. That is, the convolutional neural network extracts high-dimensional implicit features in the differential feature map. The reason is that the relatively shallow network structure is used for the processing of the image of the cross-section of the steel structure and the image of the connection area before, and the features extracted by it are closer to the surface features and cannot fully represent more abstract features.

[0068] In step S150, the depth feature map is passed through multiple fully connected layers to obtain a feature value. Among them, the output of the last fully connected layer in the multiple fully connected layers is one-dimensional. That is, the depth feature map is processed by multiple fully connected layers to make full use of the information at each position in the depth feature map to obtain the feature value. Essentially, this is a regression operation: based on the representation of the shape features of the injected space in the high-dimensional feature space, the feature value representing the tilt angle of the cartridge is regressed.

[0069] In step S160, based on the feature value, the target tilt angle of the cartridge is determined. Specifically, the process of determining the target tilt angle of the cartridge based on the feature value includes: First, normalize the feature value according to the value range of the feature value obtained during the training process. That is, normalize the feature value to the interval from 0 to 1 according to its value range obtained during the training process; then, map the normalized feature value to the interval of the minimum and maximum values of the tilt angle of the cartridge to obtain the target tilt angle of the cartridge.

[0070] Figure 4 The flowchart shows the method for controlling the tilt angle of the cartridge of the grouting device based on the set difference of shape features according to the embodiment of the present application. Based on the feature value, the target tilt angle of the cartridge is determined. As Figure 4 shown, determining the target tilt angle of the cartridge based on the feature value includes: S210, normalizing the feature value according to the value range of the feature value obtained during the training process; and S220, mapping the normalized feature value to the interval of the minimum and maximum values of the tilt angle of the cartridge to obtain the target tilt angle of the cartridge.

[0071] After obtaining the target tilt angle, further, based on the target tilt angle of the cartridge and the current tilt angle of the cartridge, the tilt control angle of the cartridge can be obtained. That is, in the technical solution of the present application, the method further includes: obtaining the current tilt angle of the cartridge; and calculating the difference between the current tilt angle and the target tilt angle to obtain the tilt control angle.

[0072] In summary, the method for controlling the inclination of the barrel of the grouting device based on the set difference of shape features according to the embodiments of the present application is elucidated. It determines the inclination angle of the barrel of the grouting device based on the set difference of the shape features of the cross-section of the steel structure and the shape features of the connection area between the steel structure and the ground in the high-dimensional feature space by a deep neural network.

[0073] Exemplary System

[0074] Figure 5 The block diagram of the barrel inclination control system of the grouting device based on the set difference of shape features according to the embodiments of the present application is illustrated.

[0075] As Figure 5 shown, the barrel inclination control system 500 of the grouting device based on the set difference of shape features according to the embodiments of the present application includes: an image acquisition unit 510 for acquiring images of the cross-section of the steel structure and the connection area between the steel structure and the ground, wherein the images of the cross-section of the steel structure and the connection area have the same scale; a convolutional feature map generation unit 520 for respectively passing the images of the cross-section of the steel structure and the connection area obtained by the image acquisition unit 510 through N convolutional layers to obtain a first convolutional feature map and a second convolutional feature map, where N is a positive integer greater than or equal to 2 and less than or equal to 3; a differential feature map generation unit 530 for calculating the difference between the first convolutional feature map and the second convolutional feature map obtained by the convolutional feature map generation unit 520 to obtain a differential feature map, and the differential feature map represents the result of the set difference operation of the shape features of the cross-section of the steel structure and the connection area in the high-dimensional feature space; a depth feature map generation unit 540 for passing the differential feature map obtained by the differential feature map generation unit 530 through a convolutional neural network to obtain a depth feature map; an eigenvalue generation unit 550 for passing the depth feature map obtained by the depth feature map generation unit 540 through a plurality of fully connected layers to obtain eigenvalues, wherein the output of the last fully connected layer in the plurality of fully connected layers is one-bit; and a target inclination angle generation unit 560 for determining the target inclination angle of the barrel based on the eigenvalues obtained by the eigenvalue generation unit 550.

[0076] In one example, in the above barrel inclination control system 500, the images of the cross-section of the steel structure and the connection area are acquired under the same shooting conditions.

[0077] In one example, in the above-mentioned barrel tilting control system 500, the image acquisition unit 510 is further configured to: acquire a first depth of field when photographing the cross-section of the steel structure and a second depth of field when photographing the connection area; and convert the image of the cross-section of the steel structure and the image of the connection area into images of the same scale based on the first depth of field or the second depth of field.

[0078] In one example, in the above-mentioned barrel tilting control system 500, the differential feature map generation unit 530 is further configured to: subtract the first convolutional feature map from the second convolutional feature map by pixel position to obtain the differential feature map.

[0079] In one example, in the above-mentioned barrel tilting control system 500, as Figure 6 shown, the target tilt angle generation unit 560 includes: a normalization processing subunit 561 configured to perform normalization processing on the eigenvalue according to the value range of the eigenvalue obtained during the training process; and a numerical mapping subunit 562 configured to map the normalized eigenvalue to the interval between the minimum value and the maximum value of the tilt angle of the barrel to obtain the target tilt angle of the barrel.

[0080] In one example, in the above-mentioned barrel tilting control system 500, the barrel tilting control system 500 further includes: a tilt control unit 570 configured to: acquire the current tilt angle of the barrel; and calculate the difference between the current tilt angle and the target tilt angle to obtain a tilt control angle.

[0081] Here, those skilled in the art can understand that the specific functions and operations of each unit and module in the above-mentioned barrel tilting control system 500 have been described in detail in the description of the barrel tilting control method of the grouting device based on the set difference of shape features above, and therefore, the repeated description thereof will be omitted. Figures 1 to 4 As described above, the barrel tilting control system 500 according to the embodiments of the present application can be implemented in various terminal devices, such as the controller of a grouting device. In one example, the barrel tilting control system 500 according to the embodiments of the present application can be integrated into the terminal device as a software module and / or a hardware module. For example, the barrel tilting control system 500 can be a software module in the operating system of the terminal device, or can be an application program developed for the terminal device; of course, the barrel tilting control system 500 can also be one of the many hardware modules of the terminal device.

[0082]

[0083] ​Alternatively, in another example, the cartridge tilt control system 500 and the terminal device may also be separate devices, and the cartridge tilt control system 500 can be connected to the terminal device through a wired and / or wireless network and transmit interaction information in accordance with a predefined data format.

[0084] Exemplary Electronic Device

[0085] Next, an electronic device according to an embodiment of the present application will be described with reference to Figure 7 FIG.

[0086] Figure 7 FIG. shows a block diagram of an electronic device according to an embodiment of the present application.

[0087] As Figure 7 shown, the electronic device 10 includes one or more processors 11 and a memory 12.

[0088] The processor 11 may be a central processing unit (CPU) or other form of processing unit having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.

[0089] The memory 12 may include one or more computer program products, and the computer program products may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory, etc. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage media, and the processor 11 may run the program instructions to implement the cartridge tilt control method based on set difference of shape features and / or other desired functions of various embodiments of the present application described above. Various contents such as a target tilt angle and a tilt control angle may also be stored in the computer-readable storage media.

[0090] In one example, the electronic device 10 may further include: an input device 13 and an output device 14, and these components are interconnected through a bus system and / or other form of connection mechanism (not shown).

[0091] The input device 13 may include, for example, a keyboard, a mouse, etc.

[0092] The output device 14 may output various information to the outside, including the target tilt angle, etc. The output device 14 may include, for example, a display, a speaker, a printer, and a communication network and its connected remote output devices, etc.

[0093] Of course, for simplicity, Figure 7 only some of the components related to the present application in the electronic device 10 are shown, and components such as buses, input / output interfaces, etc. are omitted. In addition, according to specific application scenarios, the electronic device 10 may further include any other appropriate components.

[0094] Exemplary Computer Program Product and Computer Readable Storage Medium

[0095] In addition to the above methods and devices, an embodiment of the present application may also be a computer program product, which includes computer program instructions that, when run by a processor, cause the processor to execute the steps in the method for controlling the inclination of the barrel of the grouting device based on set difference of shape features according to various embodiments of the present application described in the "Exemplary Method" section above of this specification.

[0096] The computer program product may be written in any combination of one or more programming languages for programming code to perform the operations of the embodiments of the present application. The programming languages include object-oriented programming languages such as Java, C++, etc., and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, executed as a stand-alone software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0097] Furthermore, an embodiment of the present application may also be a computer-readable storage medium, on which computer program instructions are stored, and when the computer program instructions are run by a processor, the processor is caused to execute the steps in the method for controlling the inclination of the barrel of the grouting device based on set difference of shape features according to various embodiments of the present application described in the "Exemplary Method" section above of this specification.

[0098] The computer-readable storage medium may adopt any combination of one or more readable media. The readable media may be a readable signal medium or a readable storage medium. The readable storage medium may, for example, include but is not limited to an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0099] The basic principles of the present application have been described in conjunction with specific embodiments. However, it should be noted that the advantages, benefits, effects, etc. mentioned in the present application are only examples and not limitations. It cannot be considered that these advantages, benefits, effects, etc. are essential for each embodiment of the present application. Additionally, the specific details disclosed above are only for illustrative and easy-to-understand purposes, not limitations. The above details do not limit the present application to necessarily implement using the above specific details.

[0100] The block diagrams of the devices, apparatuses, equipment, and systems involved in the present application are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open-ended terms, meaning "including but not limited to", and can be used interchangeably with each other. The word "or" and "and" used herein refer to the phrase "and / or", and can be used interchangeably with each other, unless the context clearly indicates otherwise. The word "such as" used herein refers to the phrase "such as but not limited to", and can be used interchangeably with each other.

[0101] It should also be noted that in the devices, equipment, and methods of the present application, each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent solutions of the present application.

[0102] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of the present application. Therefore, the present application is not intended to be limited to the aspects shown herein, but rather to the broadest scope consistent with the principles and novel features disclosed herein.

[0103] The above description has been given for purposes of illustration and description. In addition, this description is not intended to limit the embodiments of the present application to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions, and sub-combinations thereof.

Claims

1. A method for controlling the inclination of the barrel of a grouting device based on set difference of shape features, characterized in that, Including: Obtain images of the cross-section of the steel structure and the connection area between the steel structure and the ground, wherein the images of the cross-section of the steel structure and the connection area have the same scale; Respectively pass the images of the cross-section of the steel structure and the connection area through N convolutional layers to obtain a first convolutional feature map and a second convolutional feature map, where N is a positive integer greater than or equal to 2 and less than or equal to 3; Calculate the difference between the first convolutional feature map and the second convolutional feature map to obtain a difference feature map, and the difference feature map represents the representation after performing set difference operation on the shape features of the cross-section of the steel structure and the connection area in the high-dimensional feature space; Pass the difference feature map through a convolutional neural network to obtain a depth feature map; Pass the depth feature map through multiple fully connected layers to obtain eigenvalues, where the output of the last fully connected layer in the multiple fully connected layers is one bit; and Based on the eigenvalues, determine the target tilt angle of the barrel, including: normalizing the eigenvalues according to the value range of the eigenvalues obtained during the training process; and mapping the normalized eigenvalues to the interval of the minimum and maximum values of the tilt angle of the barrel to obtain the target tilt angle of the barrel.

2. The method for controlling the inclination of the barrel of the grouting device based on the set difference of shape features according to claim 1, wherein, In obtaining the images of the cross-section of the steel structure and the connection area between the steel structure and the ground, the images of the cross-section of the steel structure and the connection area are collected under the same shooting conditions.

3. The method for controlling the inclination of the barrel of the grouting device based on the set difference of shape features according to claim 1, wherein, Obtain images of the cross-section of the steel structure and the connection area between the steel structure and the ground, including: Obtain the first depth of field when shooting the cross-section of the steel structure and the second depth of field when shooting the connection area; and Based on the first depth of field or the second depth of field, convert the images of the cross-section of the steel structure and the connection area into images of the same scale.

4. The method for controlling the inclination of the barrel of the grouting device based on the set difference of shape features according to claim 1, wherein, Calculate the difference between the first convolutional feature map and the second convolutional feature map to obtain a difference feature map, including: Subtract the first convolutional feature map from the second convolutional feature map pixel by pixel to obtain the difference feature map.

5. The method for controlling the tilt of the barrel of the grouting device based on set difference of shape features according to claim 4, further includes: Obtain the current tilt angle of the barrel; And Calculate the difference between the current tilt angle and the target tilt angle to obtain a tilt control angle.

6. A barrel inclination control system for a grouting device based on set difference of shape features, characterized in that, Including: An image acquisition unit for obtaining images of the cross-section of the steel structure and the connection area between the steel structure and the ground, wherein the images of the cross-section of the steel structure and the connection area have the same scale; A convolutional feature map generation unit for respectively passing the images of the cross-section of the steel structure and the connection area obtained by the image acquisition unit through N convolutional layers to obtain a first convolutional feature map and a second convolutional feature map, where N is a positive integer greater than or equal to 2 and less than or equal to 3; A differential feature map generation unit, configured to calculate the difference between the first convolutional feature map and the second convolutional feature map obtained by the convolutional feature map generation unit, so as to obtain a differential feature map, where the differential feature map represents the representation after the set difference operation of the shape features of the cross-section of the steel structure and the connection area in the high-dimensional feature space; A depth feature map generation unit, configured to obtain a depth feature map by passing the differential feature map obtained by the differential feature map generation unit through a convolutional neural network; An eigenvalue generation unit, configured to obtain eigenvalues by passing the depth feature map obtained by the depth feature map generation unit through a plurality of fully-connected layers, where the output of the last fully-connected layer in the plurality of fully-connected layers is one-bit; and A target tilt angle generation unit, configured to determine the target tilt angle of the cartridge based on the eigenvalues obtained by the eigenvalue generation unit, including: normalizing the eigenvalues according to the value range of the eigenvalues obtained during the training process; and mapping the normalized eigenvalues to the minimum and maximum value intervals of the tilt angle of the cartridge to obtain the target tilt angle of the cartridge.

7. The barrel tilting control system of the grouting device based on set difference of shape features according to claim 6, wherein, Images of the cross-section of the steel structure and the connection area are collected under the same shooting conditions.

8. The barrel inclination control system of the grouting device based on set difference of shape features according to claim 6, wherein, The image acquisition unit is further configured to: acquire a first depth of field when shooting the cross-section of the steel structure and a second depth of field when shooting the connection area; and convert the images of the cross-section of the steel structure and the connection area into images of the same scale based on the first depth of field or the second depth of field.

9. An electronic device, comprising: A processor; And A memory, in which computer program instructions are stored, and when the computer program instructions are run by the processor, the processor is caused to execute the method for controlling the tilt of the cartridge of the grouting device based on the set difference of shape features according to any one of claims 1-5.

Images