Method and device for identifying recovery integrity of safety pin and airspeed sleeve of undercarriage

Abstract

The invention relates to the technical field of aviation safety, in particular to an undercarriage safety pin and airspeed sleeve recovery integrity recognition method and device, and the recognition method comprises the steps: obtaining a real-time image of the interior of a tool box; the tool box is used for placing a plurality of safety pins and a plurality of airspeed sleeves recovered from the undercarriage, and a plurality of first areas for placing the safety pins and a plurality of second areas for placing the airspeed sleeves are arranged in the tool box; based on a target detection model, identifying whether each first area in the real-time image has a safety pin and whether each second area has an airspeed sleeve, and if yes, determining that all safety pins and airspeed sleeves are completely recovered; wherein the target detection model is a deep learning model which is obtained through a pre-training process and has adaptive weight parameters. According to the invention, the recovery integrity of the undercarriage safety pin and the airspeed sleeve can be accurately identified in real time.

Description

Technical Field

[0001] The invention relates to the field of aviation safety technology, and in particular to a recovery integrity identification method and device for a landing gear safety pin and a pitot tube sleeve. Background Art

[0002] In the civil aviation maintenance industry, maintenance personnel must remove the landing gear safety pin and pitot tube sleeve during each aircraft's pre-flight preparations. If the safety pin or pitot tube sleeve is not removed, serious errors will occur after the aircraft takes off. The current method for detecting whether the landing gear safety pin and pitot tube sleeve are completely recovered is the manual visual method. The maintenance personnel remove the landing gear safety pin and pitot tube sleeve, take photos and send them to the work group. The on-duty cadres or other designated personnel review the photos and manually check the integrity of the safety pin and pitot tube sleeve. The disadvantage of this method is that the manual review workload is large, time-consuming and labor-intensive, and the review speed is slow; moreover, manual review is prone to omissions or negligence, and no abnormality in the photos is found or the review is missed, resulting in the aircraft taking off with the safety pin or pitot tube sleeve, causing safety hazards. Summary of the invention

[0003] 1. Technical issues to be resolved

[0004] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a method and device for identifying the recovery integrity of a landing gear safety pin and a pitot tube sleeve, which solves the technical problems of large workload and easy omissions in the manual visual method.

[0005] (II) Technical solution

[0006] In order to achieve the above object, the main technical solutions adopted by the present invention include:

[0007] In a first aspect, an embodiment of the present invention provides a method for identifying the recovery integrity of a landing gear safety pin and a pitot tube sleeve, comprising: acquiring a real-time image inside a tool box; the tool box is used to place a plurality of safety pins and a plurality of pitot tube sleeves recovered from the landing gear, and the tool box is provided with a plurality of first areas designated for placing safety pins, and a second area designated for placing pitot tube sleeves;

[0008] Based on the target detection model, identifying whether there is a safety pin in each first area in the real-time image, and whether there is a pitot tube sleeve in each second area, and if so, determining that all safety pins and pitot tubes have been completely recovered;

[0009] Among them, the target detection model is a deep learning model with adapted weight parameters obtained through a pre-training process.

[0010] Optionally, the target detection model is obtained by replacing the convolution module, the CSPLayer_2Conv module and the SPPF module in the backbone network of the YOLOv8n neural network model with an inverted residual module;

[0011] Among them, the inverted residual module is a module in the MobileNetV3 neural network model.

[0012] Optionally, the input end of the backbone network of the target detection model is a Conv_BN_hswish module;

[0013] The Conv_BN_hswish module includes a common convolution unit, a batch normalization unit and an h-swish activation function unit connected sequentially.

[0014] Optionally, the activation function in the h-swish activation function unit is:

[0015]

[0016] Wherein, x represents the pixel value of each pixel in the real-time image, ReLU6=(ReLU,6), ReLU=max(0,x).

[0017] Optionally, the number of the first areas is three, the number of the second areas is three, and the pre-training process includes:

[0018] S1. Obtain at least 64 initial training images;

[0019] In the initial training image, 64 placement methods are classified according to whether there is a safety pin in each first area and whether there is a pitot tube sleeve in each second area;

[0020] S2, labeling and enhancing the safety pin and the pitot tube sleeve in the initial training image to obtain a final training image;

[0021] S3. Among all the final training images, the final training images with all the safety pins and pitot tube sleeves are selected as the positive example set, and the remaining final training images are selected as the negative example set, and the positive example set and the negative example set together constitute the training data set;

[0022] S4. Use the training data set to train the initial target detection model to obtain a target detection model with adapted weight parameters.

[0023] Optionally, the S1 includes:

[0024] According to multiple shooting angles, shooting directions and shooting brightness, a camera device is used to obtain initial training images based on 64 placement methods.

[0025] Optionally, in S2,

[0026] The labeling process includes: labeling the safety pin and the pitot tube sleeve in the initial training image using labelme software to generate a data file in JSON format;

[0027] The enhancement processing includes: cropping, rotating, translating, changing brightness, introducing noise, HSV enhancement, Cutout, flipping, introducing motion blur and introducing speckle noise.

[0028] Optionally, the S4 includes:

[0029] The training data set is divided into a training set, a validation set, and a test set in a ratio of 8:1:1;

[0030] The initial target detection model is trained using the training set, and after each round of training, the model parameters of the current target detection model are verified and updated using the validation set;

[0031] End the current training process after 300 rounds of training;

[0032] Determine whether the performance index reaches a preset value, and if so, use the current target detection model as a target detection model with adapted weight parameters;

[0033] The performance indicators include:

[0034]

[0035] Where P represents precision, TP represents the number of images correctly detected in the positive set, FP represents the number of images not correctly detected in the positive set, FN represents the number of images in the negative set that are incorrectly detected as positive, R represents the recall rate, P(R) represents the relationship curve between precision P and recall rate R, AP represents average precision, mAP represents the average value of the average precision AP of all placement classification categories, n represents the classification category, AP i Represents the average precision of the i-th classification.

[0036] In a second aspect, an embodiment of the present invention provides a recovery integrity identification device for a landing gear safety pin and a pitot tube sleeve, including a processing device;

[0037] The processing device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, the method for identifying the recovery integrity of the landing gear safety pin and pitot tube sleeve described in the first aspect is implemented.

[0038] Optionally, an interactive device is also included;

[0039] The interactive device is used to interact with the user, obtain the real-time image uploaded by the user, send the real-time image to the processing device to identify the recovery integrity of the landing gear safety pin and the pitot tube sleeve in the real-time image and obtain the identification result; and receive and display the identification result returned by the processing device to the user.

[0040] (III) Beneficial effects

[0041] The beneficial effects of the present invention are as follows: the recovery integrity identification method and device of the landing gear safety pin and pitot tube sleeve of the present invention, after acquiring the real-time image inside the tool box, based on the target detection model, according to the first area in the tool box where the safety pin is placed and the second area where the pitot tube sleeve is placed, identifies whether there is a safety pin in each first area in the real-time image and a pitot tube sleeve in each second area, and determines that all safety pins and pitot tubes have been completely recovered. Compared with the prior art, after acquiring the real-time image, the target detection model can be used to complete the remaining review work, effectively reducing the manual workload, and real-time and accurately identifying the recovery integrity of the landing gear safety pin and pitot tube sleeve, ensuring that the safety pin and pitot tube sleeve are removed and recovered into the tool box before the aircraft takes off, ensuring the safe flight of the aircraft. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 A schematic flow chart of a method for identifying the recovery integrity of a landing gear safety pin and a pitot tube sleeve provided in an embodiment;

[0043] Figure 2 It is a schematic diagram of the structure of the existing YOLOv8n neural network model;

[0044] Figure 3 A schematic diagram of the structure of the backbone network of the target detection model provided in the embodiment;

[0045] Figure 4 A schematic diagram of the structure of the inverted residual module provided in the embodiment;

[0046] Figure 5 A schematic diagram of a flow chart of a pre-training process provided in an embodiment;

[0047] Figure 6 An initial training image partially collected for the embodiment;

[0048] Figure 7 It is a line chart of the performance of the target detection model on the training set and the validation set during the training process;

[0049] Figure 8 A schematic diagram of an interface for displaying recognition results of an interactive device provided in an embodiment. DETAILED DESCRIPTION

[0050] In order to better explain the present invention and facilitate understanding, the present invention is described in detail below through specific implementation modes in conjunction with the accompanying drawings.

[0051] Embodiment 1

[0052] like Figure 1 As shown, this embodiment provides a method for recognizing the recovery integrity of a landing gear safety pin and a pitot tube sleeve, including steps A1 and A2, which are specifically as follows:

[0053] A1. Acquire a real-time image of the interior of a toolbox; the toolbox is used to place a plurality of safety pins and a plurality of pitot tube sleeves recovered from a landing gear, and the toolbox is provided with a plurality of first areas designated for placing safety pins and a second area designated for placing pitot tube sleeves.

[0054] A2. Based on the target detection model, identify whether there is a safety pin in each first area in the real-time image, and whether there is a pitot tube sleeve in each second area. If so, determine that all safety pins and pitot tubes have been completely recovered; wherein the target detection model is a deep learning model with adaptive weight parameters obtained through a pre-training process.

[0055] The recovery integrity identification method of the landing gear safety pin and pitot tube sleeve provided in this embodiment, after acquiring the real-time image inside the tool box, based on the target detection model, according to the first area designated for placing the safety pin and the second area designated for placing the pitot tube sleeve in the tool box, identifies whether there is a safety pin in each first area in the real-time image and a pitot tube sleeve in each second area, and determines that all safety pins and pitot tubes have been completely recovered. Compared with the prior art, after acquiring the real-time image, the target detection model can be used to complete the remaining review work, effectively reducing the manual workload, and real-time and accurate identification of the recovery integrity of the landing gear safety pin and pitot tube sleeve, ensuring that the safety pin and pitot tube sleeve are removed and recovered into the tool box before the aircraft takes off, ensuring the safe flight of the aircraft.

[0056] In a preferred implementation of this embodiment, in order to further improve the recognition speed and reduce the computing power requirements of the device, the target detection model is obtained by replacing the convolution module, CSPLayer_2Conv module and SPPF module in the backbone network of the YOLOv8n neural network model with an inverted residual module; wherein the inverted residual module is a module in the MobileNetV3 neural network model.

[0057] The YOLOv8n neural network model itself has a good speed advantage in the field of image detection, and has better detection performance in application scenarios with high real-time requirements. It usually includes a backbone network (backbone), a neck network (neck) and a head network (head). The structure of the existing YOLOv8n neural network model is as follows: Figure 2 shown.

[0058] This embodiment improves the backbone network of the existing YOLOv8n neural network model. Figure 3 As shown, the inverted residual module is used to replace the convolution module (ConvModule), CSPLayer_2Conv module and SPPF module in the backbone network of the YOLOv8n neural network model.

[0059] The structure of the inverted residual module is as follows Figure 4 As shown in the figure, it is a key component in the design of lightweight convolutional neural networks. It is innovatively adjusted based on the traditional residual block to adapt to the environment with limited computing resources. In the standard residual structure, the input feature map undergoes a sequence of convolutions and is added to the original input to produce the output; the inverted residual adopts a different strategy: first, the number of channels of the input feature map is expanded using an expanded convolution with a small number of output channels to enrich the feature representation, and then a depthwise separable convolution is used. This operation is performed in two steps - the depthwise convolution reduces the computational burden and is implemented through channel-by-channel convolution; the subsequent point convolution not only restores the number of channels, but also embeds nonlinear characteristics to enhance the expression ability of the model. This design exquisitely balances computational efficiency and feature extraction capabilities, which is particularly effective in lightweight networks. It can effectively reduce the computing power requirements of the device and improve the recognition speed of the target detection model.

[0060] Specifically, the inverted residual module includes a pointwise convolution submodule, a depthwise separable convolution submodule and a residual connection submodule. The working principle of the inverted residual module is as follows:

[0061] Point-wise convolution submodule: First, the input feature map is transformed through a 1×1 convolution layer. Usually this step increases the number of channels. This is where the term “inverted” comes from, because this is the opposite of the traditional convolution operation that reduces the number of channels.

[0062] Depthwise Convolution: Next, a filter is applied independently to each input channel through depthwise convolution, which greatly reduces the number of parameters. Subsequently, the channels are fused again through the pointwise convolution submodule, usually reducing the number of channels so that the number of output channels is equal to or close to the number of input channels.

[0063] Residual connection submodule: After the depthwise separable convolution submodule, if the input and output have the same shape, the input can be directly added to the output through the residual connection submodule, which helps the propagation of gradients and also enhances the expressiveness of the object detection model.

[0064] In addition, the activation function in the inverted residual module also uses the h-swish activation function, which enables the target detection model to provide better performance while maintaining computational efficiency; the expansion rate of the inverted residual module is no longer fixed, but can be adjusted according to needs, which increases the flexibility and efficiency of the model architecture; in addition, the SE module is introduced between the above sub-modules, which is an attention mechanism that can adaptively recalibrate channel weights to enhance the expressiveness of the model.

[0065] In contrast, the classic ResNet residual structure first reduces the number of channels through 1×1 convolution, then performs 3×3 convolution processing, and finally expands the channels through 1×1 convolution again, forming a "bottleneck" shape. The inverted residual method does the opposite, initially using 1×1 convolution to increase the number of channels, then applying 3×3 deep convolution (DW convolution) for efficient feature extraction, and finally reducing the channels back to the initial scale. This series of operations not only improves the model efficiency, but also maintains a strong feature learning capability.

[0066] More preferably, the input end of the backbone network of the target detection model is a Conv_BN_hswish module.

[0067] The Conv_BN_hswish module includes a common convolution unit, a batch normalization unit and an h-swish activation function unit connected sequentially.

[0068] The activation function in the h-swish activation function unit is:

[0069]

[0070] Wherein, x represents the pixel value of each pixel in the real-time image, ReLU6=(ReLU,6), ReLU=max(0,x).

[0071] Through the above-mentioned improvements to the backbone network of the YOLOv8n neural network model, the final target detection model is smaller in size, has lower requirements on the computing power resources of the recognition equipment used, and has better adaptability to the application environment.

[0072] Embodiment 2

[0073] Based on the target detection model provided in Example 1, this example specifically describes the pre-training project of the target detection model.

[0074] In a specific implementation of this embodiment, the number of the first areas of the tool box is three, and the number of the second areas is three.

[0075] like Figure 5 As shown, the pre-training process includes steps S1 to S4:

[0076] S1. Obtain training images of 64 placement methods.

[0077] In the initial training image, according to whether there is a safety pin in each first area and whether there is a pitot tube sleeve in each second area, a total of 64 placement methods are included.

[0078] Specifically, during the pre-flight preparation of each aircraft, maintenance personnel must remove the landing gear safety pin and pitot tube sleeve and place them in the tool box. If the safety pin or pitot tube sleeve is not removed, serious errors will occur after the aircraft takes off. There are 3 landing gear safety pins and 3 pitot tube sleeves on each aircraft. According to the arrangement and combination of the placement of the landing gear safety pins and pitot tube sleeves, there are 64 placement methods. The landing gear safety pins and pitot tube sleeves with different placement methods are photographed and collected with a mobile phone at multiple angles, multiple directions and different degrees of brightness changes, and a total of 8,612 images are collected.

[0079] Preferably, in order to be closer to the actual scene, images taken by maintenance personnel on site can also be collected as initial training images. In images taken on site, tool boxes, safety pins, and pitot tube sleeves are usually more worn out, and their placement may be irregular, which can enrich the diversity of the data set, so that the final object detection model has better generalization performance in actual application scenarios.

[0080] S2. Annotate and enhance the safety pin and the pitot tube sleeve in the initial training image to obtain a final training image.

[0081] Specifically, the enhancement processing includes: cropping, rotation, translation, brightness change, introduction of noise, HSV enhancement, cutout, flipping, introduction of motion blur and introduction of speckle noise. Among them, cutout refers to randomly cutting off part of the sample area and filling it with 0 pixel values, and the classification result remains unchanged. The above-mentioned cropping, rotation, translation, brightness change, introduction of noise, HSV enhancement, Cutout and flipping can be implemented using the OpenCV function package, and the introduction of motion blur and introduction of speckle noise can be implemented using the Imgaug function package. After data enhancement, there are a total of 13,680 images. Some of the training images for data enhancement are as follows Figure 6 Shown

[0082] The labeling process includes: labeling the safety pin and pitot tube sleeve in the initial training image using labelme software to generate a data file in JSON format. Specifically, the safety pin in the training image can be labeled as "aqx", and the pitot tube sleeve in the image can be labeled as "gt".

[0083] To be more specific, a Python script can be used to convert the above-mentioned JSON format data file into an XML format label file, and then the XML format label file can be converted into a TXT format label file using code, wherein the TXT format label file mainly includes data of five items: "class", "x_center", "y_center", "width" and "heighth".

[0084] S3. Among the 13680 final training images, the final training images with all the safety pins and pitot tube sleeves are selected as the positive example set, and the remaining final training images are selected as the negative example set. The positive example set and the negative example set together constitute the training data set.

[0085] S4. Use the training data set to train the initial target detection model to obtain a target detection model with adapted weight parameters.

[0086] Specifically, S4 includes sub-steps S401 to S403:

[0087] S401. The training data set is randomly divided into a training set, a validation set, and a test set in a ratio of 8:1:1, and Pycharm can be used to configure yaml files of different categories of data sets.

[0088] S402: Use the training set to train the initial target detection model. After each round of training, use the validation set to validate and update the model parameters of the current target detection model.

[0089] S403: After 300 rounds of training, the current training process ends.

[0090] S404, determining whether the performance indicator reaches a preset value;

[0091] If not, jump back to S401 to continue training;

[0092] If yes, then stop training and use the current target detection model as the target detection model with adapted weight parameters. In fact, the adapted weight parameters can be exported as the best.pt weight file to facilitate migration to the recognition device together with the target detection model.

[0093] In S404, the performance indicators include:

[0094]

[0095] Where P represents precision, TP represents the number of images correctly detected in the positive set, FP represents the number of images not correctly detected in the positive set, FN represents the number of images in the negative set that are incorrectly detected as positive, R represents the recall rate, P(R) represents the relationship curve between precision P and recall rate R, AP represents average precision, mAP represents the average value of the average precision AP of all placement classification categories, n represents the classification category, AP i Represents the average precision of the i-th classification.

[0096] Figure 7 It is a performance line chart of the target detection model on the training set and the validation set during the training process from S401 to S403 above.

[0097] Based on the above training process, this embodiment also trains and compares the performance of the existing YOLOv8n neural network model and the target detection model provided in Example 1. Set the initial learning rate to 0.01, the final learning rate to 0.01, the batch (batch size, that is, the number of training images in each round of training input target detection model) to 16, and the epoch (number of cycles, that is, the number of training rounds) to 300. The configuration of this experimental environment is shown in Table 1, and the results obtained by the YOLOv8n neural network model training are shown in Table 2. The results obtained by the target detection model training provided in Example 1 are shown in Table 3. The comparison of the target detection model proposed in this embodiment with the YOLOv8n neural network model in terms of model parameter quantity, model calculation amount and model size is shown in Table 4.

[0098] Table 1 Experimental environment configuration table

[0099]

[0100]

[0101] Table 2 YOLOv8 network model training results

[0102]

[0103] Table 3. Target detection model training results provided by Example 1

[0104]

[0105] Table 4 Comparison of results

[0106]

[0107] It can be found from Tables 2 and 3 that the target detection model provided in Example 1 and the YOLOv8n neural network model have little difference in the numerical changes in the indicators of precision, recall rate, mAP50 and mAP50-95. It can be found from Table 4 that after replacing the backbone network of the YOLOv8n model with the inverted residual module in MobileNetV3, the model parameters, GFLOPs (GigaFloating-point Operations Per Second, 1 billion floating-point operations per second) and model size are significantly reduced, the parameters are reduced by 60.47%, the model calculation amount is reduced by 65.43%, and the model size is reduced by 57.45%, achieving the lightweight of the landing gear safety pin and pitot tube sleeve detection model.

[0108] Embodiment 3

[0109] Based on the identification methods provided in the first and second embodiments, this embodiment provides a recovery integrity identification device for a landing gear safety pin and a pitot tube sleeve, including a processing device and an interaction device.

[0110] The processing device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, the method for identifying the recovery integrity of the landing gear safety pin and pitot tube sleeve described in the first aspect is implemented.

[0111] Specifically, the identification result of the processing equipment may include not only "whether all safety pins and pitot tubes have been completely recovered", but also the actual number of safety pins, the actual number of pitot tube sleeves and other information, so as to provide users with more reference information and facilitate users to conduct further data statistics or processing.

[0112] The interactive device is used to interact with the user, obtain the real-time image uploaded by the user, send the real-time image to the processing device to identify the recovery integrity of the landing gear safety pin and the pitot tube sleeve in the real-time image and obtain the identification result; and receive and display the identification result returned by the processing device to the user.

[0113] In a specific implementation, an interactive terminal is installed on the interactive device, and the interactive terminal displays an interface of the recognition result such as Figure 8 As shown, in addition to marking each safety pin and pitot tube sleeve in the real-time image, the actual number of safety pins and the actual number of pitot tube sleeves are also displayed for users to review or perform other post-processing steps.

[0114] Preferably, in order to improve the portability of the interactive device, the interactive device can be a mobile phone. The maintenance personnel can use the mobile phone to take a photo, upload it to the processing device for recognition through the interactive terminal, and the interactive terminal receives the recognition result returned by the processing device and displays it to the user. In addition, in conjunction with the target detection model obtained by the training process of using the mobile phone to collect the initial training image in Example 2, the mobile phone is used as the interactive device, and the real-time image it collects is consistent with the data style of the training data set used by the target detection model, which can further ensure the recognition accuracy of the target detection model.

[0115] In the description of the present invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0116] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0117] In the present invention, unless otherwise clearly specified and limited, when a first feature is “on” or “below” a second feature, it may be that the first and second features are in direct contact, or that the first and second features are in indirect contact through an intermediate medium. Moreover, when a first feature is “above”, “above” or “above” a second feature, it may be that the first feature is directly above or obliquely above the second feature, or it may simply mean that the first feature is higher in level than the second feature. When a first feature is “below”, “below” or “below” a second feature, it may be that the first feature is directly below or obliquely below the second feature, or it may simply mean that the first feature is lower in level than the second feature.

[0118] In the description of this specification, the description of the terms "one embodiment", "some embodiments", "embodiment", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0119] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may alter, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A method for recognizing the recovery integrity of a landing gear safety pin and a pitot tube sleeve, characterized in that: include: Get a live image of the inside of the toolbox; The tool box is used to place a plurality of safety pins and a plurality of pitot tube sleeves recovered from the landing gear, and the tool box is provided with a plurality of first areas designated for placing safety pins and a second area designated for placing pitot tube sleeves; Based on the target detection model, identifying whether there is a safety pin in each first area in the real-time image, and whether there is a pitot tube sleeve in each second area, and if so, determining that all safety pins and pitot tubes are complete; Among them, the target detection model is a deep learning model with adapted weight parameters obtained through a pre-training process.

2. The identification method according to claim 1, characterized in that: The target detection model is obtained by replacing the convolution module, CSPLayer_2Conv module and SPPF module in the backbone network of the YOLOv8n neural network model with an inverted residual module; Among them, the inverted residual module is a module in the MobileNetV3 neural network model.

3. The identification method according to claim 2, characterized in that: The input end of the backbone network of the target detection model is the Conv_BN_hswish module; The Conv_BN_hswish module includes a common convolution unit, a batch normalization unit and an h-swish activation function unit connected sequentially.

4. The identification method according to claim 3, characterized in that: The activation function in the h-swish activation function unit is: Wherein, x represents the pixel value of each pixel in the real-time image, ReLU6=(ReLU,6), ReLU=max(0,x).

5. The identification method according to claim 1, characterized in that: The number of the first regions is three, the number of the second regions is three, and the pre-training process includes: S1. Obtain at least 64 initial training images; In the initial training image, 64 placement methods are classified according to whether there is a safety pin in each first area and whether there is a pitot tube sleeve in each second area; S2, labeling and enhancing the safety pin and the pitot tube sleeve in the initial training image to obtain a final training image; S3. Among all the final training images, the final training images with all the safety pins and pitot tube sleeves are selected as the positive example set, and the remaining final training images are selected as the negative example set, and the positive example set and the negative example set together constitute the training data set; S4. Use the training data set to train the initial target detection model to obtain a target detection model with adapted weight parameters.

6. The identification method according to claim 5, characterized in that: The S1 includes: According to multiple shooting angles, shooting directions and shooting brightness, a camera device is used to obtain initial training images based on 64 placement methods.

7. The identification method according to claim 5, characterized in that: In said S2, The labeling process includes: labeling the safety pin and the pitot tube sleeve in the initial training image using labelme software to generate a data file in JSON format; The enhancement processing includes: cropping, rotating, translating, changing brightness, introducing noise, HSV enhancement, Cutout, flipping, introducing motion blur and introducing speckle noise.

8. The identification method according to claim 5, characterized in that: The S4 includes: The training data set is divided into a training set, a validation set, and a test set in a ratio of 8:1:1; The initial target detection model is trained using the training set, and after each round of training, the model parameters of the current target detection model are verified and updated using the validation set; End the current training process after 300 rounds of training; Determine whether the performance index reaches a preset value, and if so, use the current target detection model as a target detection model with adapted weight parameters; The performance indicators include: Where P represents precision, TP represents the number of images correctly detected in the positive set, FP represents the number of images not correctly detected in the positive set, FN represents the number of images in the negative set that are incorrectly detected as positive, R represents the recall rate, P(R) represents the relationship curve between precision P and recall rate R, AP represents average precision, mAP represents the average value of the average precision AP of all placement classification categories, n represents the classification category, AP i Represents the average precision of the i-th classification.

9. A recovery integrity identification device for landing gear safety pin and pitot tube sleeve, characterized in that: including processing equipment; The processing device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, the method for identifying the recovery integrity of a landing gear safety pin and a pitot tube sleeve according to any one of claims 1 to 8 is implemented.

10. The identification device according to claim 9, characterized in that: It also includes interactive devices; The interactive device is used to interact with the user, obtain the real-time image uploaded by the user, send the real-time image to the processing device to identify the recovery integrity of the landing gear safety pin and the pitot tube sleeve in the real-time image and obtain the identification result; and receive and display the identification result returned by the processing device to the user.

Images