Processing method and system for high-strength light-weight airframe of unmanned aerial vehicle

Abstract

The invention discloses a machining method and system for a high-strength light-weight vehicle body of an unmanned aerial vehicle. The method comprises the steps that firstly, materials are cut according to the design requirements of the unmanned aerial vehicle to obtain cut materials, then, the cut materials are subjected to surface treatment to obtain surface-treated materials, then, the surface-treated materials are assembled according to a design structure diagram of the unmanned aerial vehicle to obtain unmanned aerial vehicle body components, and then the unmanned aerial vehicle body components are assembled according to a design structure diagram of the unmanned aerial vehicle. Detecting the unmanned aerial vehicle body component to obtain a defect detection result, and finally, responding to the defect detection result that the detected unmanned aerial vehicle body component does not have the internal structure defect, and installing the unmanned aerial vehicle body component to complete manufacturing of the unmanned aerial vehicle. In this way, the influence of internal structure defects of unmanned aerial vehicle body components can be reduced.

Description

Technical Field

[0001] The present application relates to the field of unmanned aerial vehicle processing, and more specifically, to a processing method and system for a high-strength and lightweight body of a unmanned aerial vehicle. Background Art

[0002] Drones have a wide range of applications, such as military, scientific research, rescue, entertainment, etc. The body structure of a drone is an important factor affecting its performance and safety, so high-strength and lightweight materials and specific processing methods are required to make drones.

[0003] At present, commonly used drone body materials include metal, composite materials, plastics, etc., but these materials will produce certain defects during the processing process, such as cracks, deformation, bubbles, etc., which affect the structural integrity and durability of the drone. In addition, there are also some problems with the traditional drone body processing methods, such as low cutting accuracy, uneven surface treatment, low assembly efficiency, unreliable detection methods, etc., which lead to unstable quality and performance of drones. Therefore, developing a processing method for high-strength and lightweight drone bodies is an urgent need and technical difficulty in the current drone manufacturing field. Summary of the invention

[0004] In view of this, the present application proposes a processing method and system for a high-strength and lightweight body of a UAV, which can ensure the structural integrity and durability of the UAV by performing internal structural defect detection on the UAV body components.

[0005] According to one aspect of the present application, a method for processing a high-strength and lightweight body of a drone is provided, comprising:

[0006] Cut the material according to the design requirements of the drone to obtain cut materials;

[0007] Performing surface treatment on the cut material to obtain a surface-treated material;

[0008] According to the design structure diagram of the UAV, assembling the surface treated materials to obtain UAV body parts;

[0009] Inspecting the drone body parts to obtain defect detection results; and

[0010] In response to the defect detection result that the detected drone body parts do not have internal structural defects, the drone body parts are installed to complete the production of the drone.

[0011] According to another aspect of the present application, a processing system for a high-strength and lightweight body of a drone is provided, comprising:

[0012] A cutting module, used for cutting the material according to the design requirements of the UAV to obtain cut materials;

[0013] A surface treatment module, used for performing surface treatment on the cut material to obtain a surface treated material;

[0014] An airframe component assembly module, used to assemble the surface-treated materials according to the design structure diagram of the UAV to obtain the UAV airframe components;

[0015] A defect detection module, used to detect the drone body parts to obtain defect detection results; and

[0016] The installation module is used to install the drone body parts to complete the production of the drone in response to the defect detection result that the detected drone body parts do not have internal structural defects.

[0017] According to the embodiment of the present application, the material is first cut according to the design requirements of the drone to obtain the cut material, then the cut material is surface treated to obtain the surface treated material, then the surface treated material is assembled according to the design structure diagram of the drone to obtain the drone body parts, then the drone body parts are inspected to obtain the defect detection results, and finally, in response to the defect detection result that the inspected drone body parts do not have internal structural defects, the drone body parts are installed to complete the production of the drone. In this way, the impact of the internal structural defects of the drone body parts can be reduced.

[0018] Other features and aspects of the present application will become apparent from the following detailed description of exemplary embodiments with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate exemplary embodiments, features, and aspects of the present application and, together with the description, serve to explain the principles of the present application.

[0020] Figure 1 A flow chart showing a method for processing a high-strength and lightweight body for a drone according to an embodiment of the present application.

[0021] Figure 2 A flowchart of sub-step S140 of a method for processing a high-strength and lightweight body for a drone according to an embodiment of the present application is shown.

[0022] Figure 3 A schematic diagram of the architecture of sub-step S140 of a method for processing a high-strength and lightweight body of a drone according to an embodiment of the present application is shown.

[0023] Figure 4 A flowchart of sub-step S143 of a method for processing a high-strength and lightweight body for a drone according to an embodiment of the present application is shown.

[0024] Figure 5 A flowchart of sub-step S144 of a method for processing a high-strength and lightweight body for a drone according to an embodiment of the present application is shown.

[0025] Figure 6 A block diagram of a processing system for a high-strength and lightweight body of a drone according to an embodiment of the present application is shown.

[0026] Figure 7 A diagram showing an application scenario of a method for processing a high-strength and lightweight body for a drone according to an embodiment of the present application. DETAILED DESCRIPTION

[0027] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0028] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0029] Various exemplary embodiments, features and aspects of the present application will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise specified.

[0030] In addition, in order to better illustrate the present application, numerous specific details are given in the following specific embodiments. It should be understood by those skilled in the art that the present application can also be implemented without certain specific details. In some examples, methods, means, components and circuits well known to those skilled in the art are not described in detail in order to highlight the subject matter of the present application.

[0031] The present application provides a method for processing a high-strength and lightweight body of a drone. Figure 1 A flowchart showing a method for processing a high-strength and lightweight body for a drone according to an embodiment of the present application is shown. Figure 1As shown, according to the embodiment of the present application, the processing method for a high-strength and lightweight body of a drone includes the following steps: S110, cutting the material according to the design requirements of the drone to obtain a cut material; S120, surface treating the cut material to obtain a surface-treated material; S130, assembling the surface-treated material according to the design structure diagram of the drone to obtain drone body parts; S140, inspecting the drone body parts to obtain defect detection results; and, S150, in response to the defect detection result that the inspected drone body parts do not have internal structural defects, installing the drone body parts to complete the production of the drone.

[0032] In particular, considering that the drone may be subjected to impact and vibration from various external factors during use, if the body parts have internal structural defects, such as cracks, bubbles, etc., it may lead to reduced strength of the body, or even structural damage or failure, thus affecting the performance and safety of the drone. Therefore, in step S140, it is expected that the internal structure defect detection of the drone body parts is carried out to ensure the structural integrity and durability of the drone. However, traditional defect detection methods usually require the drone body parts to be disassembled or cut, which has a certain impact and waste on the manufacturing process of the drone.

[0033] In response to the above technical problems, the technical concept of this application is to use X-ray detection, a non-destructive detection method, combined with deep learning algorithms to inspect and identify internal defects of body parts.

[0034] Based on this, Figure 2 A flowchart of sub-step S140 of a method for processing a high-strength and lightweight body for a drone according to an embodiment of the present application is shown. Figure 3 FIG. 1 is a schematic diagram showing a sub-step S140 of a method for processing a high-strength and lightweight body of a drone according to an embodiment of the present application. Figure 2 and Figure 3 As shown, according to the processing method of the high-strength and lightweight body for a drone in an embodiment of the present application, the drone body parts are inspected to obtain defect detection results, including the steps of: S141, obtaining an X-ray detection image of the drone body parts being inspected; S142, performing image denoising processing on the X-ray detection image to obtain an X-ray detection image of the body parts after denoising; S143, extracting internal structural features of the X-ray detection image of the body parts after denoising; and, S144, determining the defect detection results based on the internal structural features.

[0035] It should be understood that in step S141, an X-ray device is used to scan the drone body parts to obtain an X-ray detection image. The X-ray image can provide information about the internal structure of the body parts, including possible defects or abnormalities. In step S142, the X-ray image may be affected by noise, which will interfere with the accurate detection and analysis of defects. Therefore, in this step, the X-ray detection image is subjected to image denoising to remove the noise in the image and obtain the X-ray detection image of the body parts after denoising, which can improve the accuracy and reliability of subsequent steps. In step S143, internal structural features are extracted from the X-ray detection image of the body parts after denoising. These features may include information such as edges, textures, shapes, etc., which are used to describe the internal structure of the body parts. Extracting these features helps to detect and analyze defects in the body parts in subsequent steps. In step S144, defect detection and analysis are performed based on the extracted internal structural features. By comparing with the pre-defined defect pattern or feature, it can be determined whether the body parts have defects. The purpose of this step is to determine whether the body parts are normal or whether there are any defects or abnormalities based on the internal structural features. In summary, these steps constitute a process for defect detection of drone body parts. By acquiring X-ray detection images, noise reduction, extracting internal structural features, and determining defect detection results based on these features, defect detection and analysis of drone body parts can be performed. This process can help improve the safety and reliability of drones and reduce potential failure risks.

[0036] Specifically, in the technical solution of the present application, an X-ray detection image of the detected drone body component is first obtained; and the X-ray detection image is subjected to image noise reduction processing to obtain the X-ray detection image of the body component after noise reduction. It should be understood that in the actual process of collecting X-ray detection images, random noise may be generated due to the instability of the X-ray source or the sensitivity of the detector, and this noise will make the X-ray detection image unclear; in addition, since X-rays may be affected by physical effects such as scattering or absorption when penetrating body components, causing non-uniform noise, this noise will cause artifacts such as stripes, spots or shadows to appear in the X-ray detection image, interfering with the real information of the image. Image noise reduction processing can reduce the noise of the X-ray detection image to a certain extent, thereby restoring its ability to express the internal structural information of the drone body components.

[0037] Generally speaking, the internal structural defects of body parts (such as cracks, bubbles, etc.) are usually manifested as low-density areas in X-ray detection images, which can form a clear contrast with the surrounding high-density areas; in addition, since the internal structural defects have a certain shape, size and direction in the X-ray detection image, the internal structural defects of the body parts can be inferred through the geometric information in the X-ray detection image of the body parts after denoising. In other words, when there are internal structural defects in the body parts of the drone, they will be manifested in the X-ray detection image. Therefore, in the technical solution of the present application, the X-ray detection image of the body parts after denoising is passed through an internal structure image feature extractor based on a convolutional neural network model to obtain an internal structure X-ray detection feature map. In this way, the internal structural features implied in the X-ray detection image of the body parts after denoising are extracted to reflect the internal structural state and performance of the body parts, as well as the potential risks and hazards that may exist.

[0038] Accordingly, in step S143, if Figure 4 As shown, extracting the internal structural features of the X-ray detection image of the body component after denoising includes: S1431, passing the X-ray detection image of the body component after denoising through an internal structural image feature extractor based on a convolutional neural network model to obtain an internal structural X-ray detection feature map; and, S1432, using the internal structural X-ray detection feature map as the internal structural feature. It is worth mentioning that the convolutional neural network (CNN) is a deep learning model mainly used to process data with a grid structure. The core idea of ​​CNN is to extract the features of input data through components such as convolutional layers, pooling layers and fully connected layers. The convolutional layer performs a convolution operation on the input image using a set of learnable filters (convolution kernels) to capture the local features of the image. The pooling layer is used for downsampling to reduce the data dimension while retaining important features. The fully connected layer maps the extracted features to the output category. In terms of extracting internal structural features, the convolutional neural network can learn local texture, edge, shape and other features in the image, so as to distinguish and identify different structures. By training a large amount of image data, CNN can learn higher-level feature representations, allowing the model to understand the semantic information in the image. In extracting the internal structural features of the X-ray detection image of the body parts after denoising, the internal structure image feature extractor based on the convolutional neural network model can learn the internal structure features in the image by training a CNN model. The model can be trained on a large number of labeled X-ray images so that it can automatically extract features related to the internal structure. These features can be used for subsequent tasks such as image classification, target detection, segmentation, etc., or as input for feature representation of other machine learning models.

[0039] Then, the internal structure X-ray detection feature map is passed through a feature autocorrelation association enhancement module to obtain an autocorrelation enhanced internal structure X-ray detection feature map. The feature autocorrelation association enhancement module effectively aggregates the complete information of the target by constructing a similarity association between feature values ​​in the internal structure X-ray detection feature map, that is, the important feature areas in the feature distribution that characterize the internal structure defects are visualized and enhanced to guide the model to enhance the feature expression of the autocorrelation enhanced internal structure X-ray detection feature map for the internal structure defects. Subsequently, the autocorrelation enhanced internal structure X-ray detection feature map is passed through a classifier to obtain a classification result, which is used to indicate whether the drone body parts being inspected have internal structure defects.

[0040] Accordingly, in step S144, if Figure 5 As shown, based on the internal structure features, the defect detection result is determined, including: S1441, the internal structure X-ray detection feature map is passed through the feature autocorrelation association enhancement module to obtain the autocorrelation enhanced internal structure X-ray detection feature map; and, S1442, the autocorrelation enhanced internal structure X-ray detection feature map is passed through the classifier to obtain a classification result, and the classification result is used to indicate whether the detected drone body component has an internal structure defect. It should be understood that in step S1441, the internal structure X-ray detection feature map is processed by the feature autocorrelation association enhancement module to obtain the autocorrelation enhanced internal structure X-ray detection feature map. The function of the feature autocorrelation association enhancement module is to enhance the correlation between features, so as to better capture the correlation information between internal structure features, so as to improve the performance of the subsequent classifier, so that it can more accurately judge the existence of defects. In step S1442, the internal structure X-ray detection feature map that has been enhanced by the feature autocorrelation association is input into the classifier to obtain a classification result. The classifier is a trained machine learning model that can classify images according to the input feature map. Here, the task of the classifier is to determine whether the drone body parts under inspection have internal structural defects. The classification results can be expressed as a binary classification of defective or non-defective, or a multivariate classification of more categories, depending on the specific problem setting. Through the processing of the above two steps, the internal structural features can be used to determine whether the drone body parts have internal structural defects. The combination of feature autocorrelation association enhancement module and classifier can improve the accuracy and reliability of defect detection, thereby providing important support for the safe operation of drones.

[0041] Among them, in step S1441, the internal structure X-ray detection feature map is passed through a feature autocorrelation association enhancement module to obtain an autocorrelation enhanced internal structure X-ray detection feature map, including: passing the internal structure X-ray detection feature map through a first convolution layer to obtain a reduced dimension feature map; passing the reduced dimension feature map through a second convolution layer to obtain an efficient associated structure map; using a cosine similarity operation to calculate a relationship matrix of the efficient associated structure map; using a Softmax function to normalize the relationship matrix to obtain a normalized relationship matrix; and using an element-by-element multiplication operation to complete The normalized relationship matrix models the relationship between any two eigenvalues ​​in the dimensionality reduction feature map to obtain a correlation feature map; deconvolution is performed on the correlation feature map to obtain a deconvolution correlation feature map; the deconvolution correlation feature map and the dimensionality reduction feature map are element-by-element added to obtain a preliminary result feature map; and, after channel expansion is performed on the preliminary result feature map to obtain an expanded preliminary result feature map, the expanded preliminary result feature map and the internal structure X-ray detection feature map are residually connected to obtain the autocorrelation enhanced internal structure X-ray detection feature map.

[0042] Among them, in step S1442, the autocorrelation enhanced internal structure X-ray detection feature map is passed through a classifier to obtain a classification result, and the classification result is used to indicate whether the inspected drone body component has internal structural defects, including: expanding the autocorrelation enhanced internal structure X-ray detection feature map into a classification feature vector according to a row vector or a column vector; using the fully connected layer of the classifier to fully connect encode the classification feature vector to obtain an encoded classification feature vector; and inputting the encoded classification feature vector into the Softmax classification function of the classifier to obtain the classification result.

[0043] That is, in the technical solution of the present application, the labels of the classifier include the presence of internal structural defects in the detected drone body parts (first label), and the absence of internal structural defects in the detected drone body parts (second label), wherein the classifier determines which classification label the autocorrelation enhanced internal structure X-ray detection feature map belongs to by means of a soft maximum function. It is worth noting that the first label p1 and the second label p2 here do not contain artificially set concepts. In fact, during the training process, the computer model does not have the concept of "whether the detected drone body parts have internal structural defects". It only has two classification labels and the probability of the output feature under these two classification labels, that is, the sum of p1 and p2 is one. Therefore, the classification result of whether the detected drone body parts have internal structural defects is actually converted into a binary classification class probability distribution that conforms to natural laws through the classification labels. In essence, the physical meaning of the natural probability distribution of the labels is used, rather than the linguistic text meaning of "whether the detected drone body parts have internal structural defects".

[0044] It should be understood that the role of the classifier is to use the given categories and known training data to learn classification rules and classifiers, and then classify (or predict) unknown data. Logistic regression and SVM are often used to solve binary classification problems. For multi-class classification problems, logistic regression or SVM can also be used, but multiple binary classifications are required to form a multi-classification. However, this is prone to errors and is not efficient. Commonly used multi-classification methods include the Softmax classification function.

[0045] In the technical solution of the present application, each feature matrix of the internal structure X-ray detection feature map expresses the image semantic features of the denoised body component X-ray detection map, and the respective feature matrices follow the channel distribution of the convolutional neural network model. After passing through the feature autocorrelation association enhancement module, each feature vector along the channel of the internal structure X-ray detection feature map can be autocorrelatedly enhanced based on the distribution of image semantic features in the spatial dimension of the feature matrix, thereby improving the overall spatial distribution association effect of the image semantic features of the autocorrelation enhanced internal structure X-ray detection feature map in the channel dimension, that is, improving the expression effect of the autocorrelation enhanced internal structure X-ray detection feature map in the channel dimension.

[0046] But at the same time, this also makes the autocorrelation enhanced internal structure X-ray detection feature map have a feature representation with dense distribution dimension correlation in the overall feature map distribution dimension, the image semantic feature space distribution in the feature matrix and the feature channel distribution between feature matrices, resulting in the probability convergence and divergence when the autocorrelation enhanced internal structure X-ray detection feature map is classified and regressed through a classifier, thereby affecting the accuracy of the classification result.

[0047] Therefore, when the present application classifies the autocorrelation enhanced internal structure X-ray detection characteristic map through a classifier, the autocorrelation enhanced internal structure X-ray detection characteristic map is first optimized, including the following steps:

[0048] Determine the median eigenvalue, the maximum eigenvalue and the minimum eigenvalue in the feature set of the autocorrelation enhanced internal structure X-ray detection feature map, and divide the median eigenvalue by the difference between the maximum eigenvalue and the minimum eigenvalue to obtain the autocorrelation enhanced internal structure X-ray detection distribution probability value, that is, p=f mid / (f max -f min ), where f mid 、f max 、f min represent the median eigenvalue, the maximum eigenvalue and the minimum eigenvalue in the feature set of the autocorrelation enhanced internal structure X-ray detection feature map respectively, and p represents the distribution probability value of the autocorrelation enhanced internal structure X-ray detection;

[0049] Each eigenvalue in the autocorrelation enhanced internal structure X-ray detection characteristic map is normalized to its maximum value to obtain the autocorrelation enhanced internal structure X-ray detection probability eigenvalue f i ;

[0050] The X-ray detection probability characteristic value f of each autocorrelation enhanced internal structure is calculated by subtracting the autocorrelation enhanced internal structure X-ray detection distribution probability value p as an exponent. i The power function of the first autocorrelation enhancement internal structure X-ray detection probability convergence eigenvalue f 1i =f i 1-p , where p represents the distribution probability value of X-ray detection of the internal structure enhanced by autocorrelation, and f i represents the characteristic value of the X-ray detection probability of the autocorrelation-enhanced internal structure, f 1i It represents the convergence eigenvalue of the probability of X-ray detection of the internal structure enhanced by the first autocorrelation;

[0051] The autocorrelation enhanced internal structure X-ray detection probability distribution value p is used as an index to calculate a probability characteristic value f of each autocorrelation enhanced internal structure X-ray detection probability characteristic value f. iThe power function of the difference of the second autocorrelation enhancement internal structure X-ray detection probability convergence eigenvalue f 2i =(1-f i ) p , where f 2i It represents the convergence eigenvalue of the probability of X-ray detection of the internal structure enhanced by the second autocorrelation;

[0052] Each of the autocorrelations enhances the internal structure X-ray detection probability characteristic value f i The first autocorrelation enhanced internal structure X-ray detection probability limited characteristic value f is multiplied by the difference between one and the autocorrelation enhanced internal structure X-ray detection distribution probability value p to obtain the first autocorrelation enhanced internal structure X-ray detection probability limited characteristic value f 3i =f i ×(1-p), where f 3i It represents the first autocorrelation enhanced internal structure X-ray detection probability limit eigenvalue;

[0053] The autocorrelation enhanced internal structure X-ray detection probability characteristic value f i The difference between the values ​​of and is multiplied by the autocorrelation enhanced internal structure X-ray detection distribution probability value p to obtain a second autocorrelation enhanced internal structure X-ray detection probability limiting characteristic value f 4i =(1-f i )×p, where f 4i It represents the limited eigenvalue of the probability of X-ray detection of the internal structure enhanced by the second autocorrelation;

[0054] After multiplying the first autocorrelation enhanced internal structure X-ray detection probability convergence eigenvalue and the second autocorrelation enhanced internal structure X-ray detection probability convergence eigenvalue, further adding the first autocorrelation enhanced internal structure X-ray detection probability limit eigenvalue and the second autocorrelation enhanced internal structure X-ray detection probability limit eigenvalue to obtain the optimized eigenvalue f' of the autocorrelation enhanced internal structure X-ray detection characteristic diagram i =f 1i ×f 2i +f 3i +f 4i .

[0055] Based on this, by taking the cross entropy form power series of the regression probability of the autocorrelation enhanced internal structure X-ray detection feature map as the probability distribution convergence limit, the regression probability convergence approximation is performed on the basis of the combination of the feature set distribution and the probability density distribution of the autocorrelation enhanced internal structure X-ray detection feature map, so as to avoid the conservative convergence control of the fatal potential divergence of the autocorrelation enhanced internal structure X-ray detection feature map, further use the probability distribution cross entropy as the objective function to guide the limited recovery strategy, so as to achieve the common agility of the convergence of the probability convergence and divergence of the feature set of the autocorrelation enhanced internal structure X-ray detection feature map to the probability density distribution space, and improve the accuracy of the classification results.

[0056] In summary, the high-strength and lightweight body processing method for a UAV based on the embodiment of the present application can ensure the structural integrity and durability of the UAV by performing internal structural defect detection on the UAV body components.

[0057] Figure 6 FIG. 1 is a block diagram of a processing system 100 for a high-strength and lightweight body of a drone according to an embodiment of the present application. Figure 6 As shown, according to the embodiment of the present application, the processing system 100 for the high-strength and lightweight body of a drone includes: a cutting module 110, which is used to cut materials according to the design requirements of the drone to obtain cut materials; a surface treatment module 120, which is used to perform surface treatment on the cut materials to obtain surface-treated materials; a body component assembly module 130, which is used to assemble the surface-treated materials according to the design structure diagram of the drone to obtain drone body components; a defect detection module 140, which is used to detect the drone body components to obtain defect detection results; and an installation module 150, which is used to install the drone body components to complete the production of the drone in response to the defect detection result that the detected drone body components do not have internal structural defects.

[0058] In a possible implementation, the defect detection module 140 includes: a detection image acquisition unit, used to acquire an X-ray detection image of the drone body component being inspected; an image denoising processing unit, used to perform image denoising on the X-ray detection image to obtain a denoised X-ray detection image of the body component; an internal structure feature extraction unit, used to extract internal structure features of the denoised X-ray detection image of the body component; and a detection result analysis unit, used to determine the defect detection result based on the internal structure features.

[0059] In a possible implementation, the internal structure feature extraction unit includes: a convolutional encoding subunit, used to pass the denoised body component X-ray detection image through an internal structure image feature extractor based on a convolutional neural network model to obtain an internal structure X-ray detection feature map; and an internal structure feature acquisition subunit, used to use the internal structure X-ray detection feature map as the internal structure feature.

[0060] Here, those skilled in the art will appreciate that the specific functions and operations of the various units and modules in the above-mentioned high-strength lightweight body processing system 100 for drones have been described in detail above. Figures 1 to 5 The processing method of the high-strength and lightweight body for a drone has been described in detail, and therefore, its repeated description will be omitted.

[0061] As described above, the processing system 100 for the high-strength and lightweight body of an unmanned aerial vehicle according to the embodiment of the present application can be implemented in various wireless terminals, such as a server having a processing algorithm for the high-strength and lightweight body of an unmanned aerial vehicle. In one possible implementation, the processing system 100 for the high-strength and lightweight body of an unmanned aerial vehicle according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the processing system 100 for the high-strength and lightweight body of an unmanned aerial vehicle can be a software module in the operating system of the wireless terminal, or can be an application developed for the wireless terminal; of course, the processing system 100 for the high-strength and lightweight body of an unmanned aerial vehicle can also be one of the many hardware modules of the wireless terminal.

[0062] Alternatively, in another example, the processing system 100 for the high-strength and lightweight body of a UAV and the wireless terminal may also be separate devices, and the processing system 100 for the high-strength and lightweight body of a UAV may be connected to the wireless terminal via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0063] Figure 7 The following is an application scenario diagram of a method for processing a high-strength and lightweight body of a drone according to an embodiment of the present application. Figure 7 As shown, in this application scenario, first, an X-ray detection image of the drone body component to be detected is obtained (for example, Figure 7 D), and then, the X-ray detection image is input to a server (for example, Figure 7As shown in S), the server is able to use the processing algorithm for the high-strength and lightweight body of the drone to process the X-ray detection image to obtain a classification result indicating whether the drone body parts being inspected have internal structural defects.

[0064] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the logical function of the specification. In some alternative implementations, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can also be executed in the opposite order sometimes, depending on the function involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be realized by a dedicated hardware-based system that performs the function or action of the specification, or can be realized by a combination of special-purpose hardware and computer instructions.

[0065] The embodiments of the present application have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A method for processing a high-strength and lightweight body of an unmanned aerial vehicle, characterized in that: include: Cut the material according to the design requirements of the drone to obtain cut materials; Performing surface treatment on the cut material to obtain a surface-treated material; According to the design structure diagram of the UAV, assembling the surface treated materials to obtain UAV body parts; Testing the drone body parts to obtain defect detection results; as well as In response to the defect detection result that the detected drone body parts do not have internal structural defects, the drone body parts are installed to complete the production of the drone.

2. The method for processing a high-strength and lightweight body for a drone according to claim 1, characterized in that: Inspecting the drone body parts to obtain defect detection results, including: Obtaining an X-ray detection image of the detected UAV body component; Performing image noise reduction processing on the X-ray detection image to obtain a noise-reduced X-ray detection image of the body component; Extracting internal structural features of the X-ray detection image of the body component after noise reduction; and Based on the internal structural features, the defect detection result is determined.

3. The method for processing a high-strength and lightweight body for a drone according to claim 2, characterized in that: Extracting the internal structural features of the X-ray detection image of the body component after noise reduction, including: Passing the de-noised X-ray detection image of the body component through an internal structure image feature extractor based on a convolutional neural network model to obtain an internal structure X-ray detection feature image; and The internal structure X-ray detection characteristic map is used as the internal structure feature.

4. The method for processing a high-strength and lightweight body for a drone according to claim 3, characterized in that: Determining the defect detection result based on the internal structure feature includes: Passing the internal structure X-ray detection characteristic map through a characteristic autocorrelation correlation enhancement module to obtain an autocorrelation enhanced internal structure X-ray detection characteristic map; and The autocorrelation enhanced internal structure X-ray detection characteristic map is passed through a classifier to obtain a classification result, and the classification result is used to indicate whether the detected drone body component has internal structural defects.

5. The method for processing a high-strength and lightweight body for a drone according to claim 4, characterized in that: The internal structure X-ray detection characteristic map is passed through a characteristic autocorrelation correlation enhancement module to obtain an autocorrelation enhanced internal structure X-ray detection characteristic map, including: Passing the internal structure X-ray detection feature map through a first convolutional layer to obtain a dimension-reduced feature map; Passing the dimension-reduced feature map through a second convolutional layer to obtain an efficient association structure map; Calculating the relationship matrix of the efficient association construction graph using a cosine similarity operation; Using a Softmax function to normalize the relationship matrix to obtain a normalized relationship matrix; The normalized relationship matrix is ​​used to model the relationship between any two eigenvalues ​​in the reduced-dimensional feature map by using an element-by-element multiplication operation to obtain a correlation feature map; Performing a deconvolution operation on the associated feature map to obtain a deconvolved associated feature map; Adding the deconvolution associated feature map and the dimension reduction feature map element by element to obtain a preliminary result feature map; and After performing channel expansion on the preliminary result feature map to obtain an expanded preliminary result feature map, the expanded preliminary result feature map and the internal structure X-ray detection feature map are residually connected to obtain the autocorrelation enhanced internal structure X-ray detection feature map.

6. The method for processing a high-strength and lightweight body for a drone according to claim 5, characterized in that: The autocorrelation enhanced internal structure X-ray detection characteristic map is passed through a classifier to obtain a classification result, and the classification result is used to indicate whether the detected drone body part has an internal structure defect, including: Expanding the autocorrelation enhanced internal structure X-ray detection feature map into a classification feature vector according to row vectors or column vectors; Performing full connection encoding on the classification feature vector using the fully connected layer of the classifier to obtain an encoded classification feature vector; and The encoded classification feature vector is input into the Softmax classification function of the classifier to obtain the classification result.

7. A high-strength and lightweight body processing system for unmanned aerial vehicles, characterized in that: include: A cutting module, used for cutting the material according to the design requirements of the UAV to obtain cut materials; A surface treatment module, used for performing surface treatment on the cut material to obtain a surface treated material; An airframe component assembly module, used to assemble the surface-treated materials according to the design structure diagram of the UAV to obtain the UAV airframe components; A defect detection module, used to detect the drone body parts to obtain defect detection results; as well as The installation module is used to install the drone body parts to complete the production of the drone in response to the defect detection result that the detected drone body parts do not have internal structural defects.

8. The high-strength and lightweight body processing system for a drone according to claim 7, characterized in that: The defect detection module comprises: A detection image acquisition unit, used to acquire an X-ray detection image of the detected UAV body component; An image noise reduction processing unit, used for performing image noise reduction processing on the X-ray detection image to obtain a noise-reduced X-ray detection image of the body component; An internal structure feature extraction unit, used to extract the internal structure features of the X-ray detection image of the body component after noise reduction; and A detection result analysis unit is used to determine the defect detection result based on the internal structure characteristics.

9. The high-strength and lightweight body processing system for a drone according to claim 8, characterized in that: The internal structure feature extraction unit comprises: a convolutional coding subunit, configured to pass the de-noised body component X-ray detection image through an internal structure image feature extractor based on a convolutional neural network model to obtain an internal structure X-ray detection feature image; and The internal structure feature acquisition subunit is used to use the internal structure X-ray detection feature map as the internal structure feature.

Images