Aircraft target detection method based on cross attention fusion feature pyramid network

Abstract

The invention discloses an aircraft target detection method based on a cross attention fusion feature pyramid network, and belongs to the technical field of remote sensing image processing. The method comprises the following steps: constructing a cross attention fusion module; building a cross attention fusion feature pyramid network based on a cross attention fusion module and a corresponding aircraft target detection model; carrying out aircraft target detection task training on the aircraft target detection model; and finally, carrying out aircraft target detection on a to-be-detected remote sensing image by utilizing the trained aircraft target detection model. According to the method, the semantic information of fusion features is enriched by effectively utilizing the advantages of the attention mechanism in capturing global semantic information association, and the multi-scale feature expression capability is improved. According to the characteristic, the model can more comprehensively extract and represent data characteristics in a remote sensing image aircraft target detection task, the image content is better understood, and therefore high-precision aircraft target detection is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of remote sensing image processing, and in particular to an aircraft target detection method based on a cross-attention fusion feature pyramid network. Background Art

[0002] In modern aviation monitoring and management systems, accurate detection of aircraft targets plays a vital role in ensuring air traffic safety, implementing border defense, and air traffic management. Traditional aircraft detection technologies mainly rely on manual interpretation of high-resolution remote sensing images and radar signal recognition, which face significant challenges in terms of high labor costs, low processing efficiency, and limited accuracy. In particular, when processing large-scale remote sensing data, traditional methods often have difficulty accurately identifying small or long-distance aircraft targets due to resolution limitations and visual complexity. In recent years, although deep learning-based target detection technologies have made remarkable achievements in fields such as autonomous driving and medical image analysis, these technologies still face certain limitations in the application of aircraft detection in remote sensing images. In particular, traditional convolutional neural networks (CNNs) and feature fusion methods, although they can extract rich features from remote sensing images, still have room for improvement in performance when dealing with aircraft scale diversity and complex backgrounds in images. These methods often ignore the interaction between features of different scales, resulting in inaccurate detection of small or heavily occluded aircraft targets in practical applications, and are easily disturbed by background noise.

[0003] Therefore, it is urgent to design a suitable technology to fully integrate the features of different scales, optimize the information flow and fusion strategy between features, explore the dependency relationship between features of different scales, enhance the ability to capture the key features of the target, and effectively improve the detection model's recognition accuracy and robustness for aircraft. Summary of the invention

[0004] In view of this, the present invention provides an aircraft target detection method and system based on a cross-attention fusion feature pyramid network that can solve the above-mentioned technical problems, which is conducive to improving the accuracy and robustness of aircraft target detection in remote sensing images.

[0005] To achieve the above object, the technical solution adopted by the present invention is:

[0006] In a first aspect, an embodiment of the present invention provides an aircraft target detection method based on a cross-attention fusion feature pyramid network, the method comprising the following steps:

[0007] S1, build cross attention fusion module;

[0008] S2. Build a cross-attention fusion feature pyramid network in combination with the cross-attention fusion module;

[0009] S3. Building an aircraft target detection model based on the cross-attention fusion feature pyramid network;

[0010] S4, training the aircraft target detection model; and performing aircraft target detection on the remote sensing image to be measured using the trained aircraft target detection model.

[0011] Furthermore, in step S1, the cross-attention fusion module constructed includes: a layer normalization module, a multi-head attention module and a multi-layer perceptron module; the cross-attention fusion module takes feature maps of adjacent scales as input, performs attention calculation on the feature maps of adjacent scales, and obtains feature dependencies between different scales.

[0012] Furthermore, in the multi-head attention module, each attention head performs attention calculation independently, and multiple attention heads are processed in parallel.

[0013] Furthermore, in the step S2, in the cross-attention fusion feature pyramid network constructed, a multi-scale feature map is used as input, and the dimensions of feature maps of different scales are adjusted through convolution operations. The adjusted feature maps are subjected to feature fusion between adjacent feature maps through a cross-attention fusion module, and then a multi-scale fused feature map is obtained through a convolution layer.

[0014] Furthermore, in step S3, the aircraft target detection model constructed includes: an image preprocessing network, a feature extraction network, a cross-attention fusion feature pyramid network and a detection head connected in sequence, wherein:

[0015] The image preprocessing network performs image preprocessing operations on remote sensing images; the preprocessed image data is sent to the feature extraction network for multi-scale feature extraction; the obtained multi-scale features are sent to the cross-attention fusion feature pyramid network to perform cross-attention feature fusion calculations to obtain a multi-scale fusion feature map; the multi-scale fusion feature map is input into the detection head to locate and classify aircraft targets.

[0016] Furthermore, the image preprocessing operations of the image preprocessing network include: horizontal flipping, noise injection, image sharpening and blurring;

[0017] The detection head includes a regression network and a classification network. The detection head uses the regression network to regress the anchor frames in the feature maps of each level based on the predefined anchor frames to accurately determine the position of the aircraft detection frame; at the same time, the classification network performs classification operations on all feature points in the feature map to evaluate the category probability of each detection frame.

[0018] Furthermore, in step S4, the model is trained for the remote sensing image aircraft target detection task using a cross entropy loss function and a mean square error loss function.

[0019] In a second aspect, an embodiment of the present invention further provides an aircraft target detection system based on a cross-attention fusion feature pyramid network, which applies the above-mentioned aircraft target detection method based on a cross-attention fusion feature pyramid network to achieve accurate aircraft target detection task recognition, and the system includes:

[0020] A target detection model building module is used to build a cross-attention fusion module, build a cross-attention fusion feature pyramid network in combination with the cross-attention fusion module, and build an aircraft target detection model based on the cross-attention fusion feature pyramid network;

[0021] A target detection model training module, used for training the aircraft target detection model for an aircraft target detection task;

[0022] The target detection model application module is used to use the trained aircraft target detection model to perform aircraft target detection on the remote sensing images to be tested.

[0023] Compared with the prior art, the present invention has at least the following beneficial effects:

[0024] 1. The present invention provides an aircraft target detection method based on a cross-attention fusion feature pyramid network, wherein a cross-attention fusion module is constructed, and a cross-attention fusion feature pyramid network and its corresponding aircraft target detection model are constructed in combination with the cross-attention fusion module. The aircraft target detection model is used to perform aircraft target detection in remote sensing images, which facilitates the realization of full-process automated aircraft target detection without manual intervention, and improves the accuracy and robustness of aircraft target detection in remote sensing images.

[0025] 2. The aircraft target detection model based on the cross-attention fusion feature pyramid network proposed in this invention enriches the semantic information of the fusion features and improves the multi-scale feature expression capability by effectively utilizing the advantage of the attention mechanism in capturing the global semantic information association. This feature enables the model to more comprehensively extract and characterize data features in the remote sensing image aircraft target detection task, better understand the image content, and thus achieve high-precision aircraft target detection.

[0026] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0027] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0029] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0030] Figure 1 A flowchart of an aircraft target detection method based on a cross-attention fusion feature pyramid network is provided in an embodiment of the present invention.

[0031] Figure 2 A schematic diagram of the multi-head attention structure provided for an embodiment of the present invention.

[0032] Figure 3 A schematic diagram of a cross-attention fusion module provided for an embodiment of the present invention.

[0033] Figure 4 Schematic diagram of the cross-attention fusion feature pyramid network provided for an embodiment of the present invention.

[0034] Figure 5 A schematic diagram of an aircraft target detection model based on a cross-attention fusion feature pyramid network provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments.

[0036] In the description of the present invention, it should be noted that: in some processes described in the specification and drawings of this application, multiple operations appearing in a specific order are included, but it should be clearly understood that these operations may not be performed in the order in which they appear in this document or may be performed in parallel. In addition, various serial numbers are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0037] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] See also Figure 1 As shown, the present invention provides an aircraft target detection method based on a cross-attention fusion feature pyramid network, which mainly includes the following steps:

[0039] S1, build cross attention fusion module;

[0040] S2. Combine the cross-attention fusion module to build a cross-attention fusion feature pyramid network;

[0041] S3. Build an aircraft target detection model based on cross-attention fusion feature pyramid network;

[0042] S4. Train the aircraft target detection model; use the trained aircraft target detection model to perform aircraft target detection on the remote sensing image to be tested.

[0043] The specific implementation methods of the method of the present invention are described in detail below:

[0044] In an embodiment of the present invention, an efficient cross-attention fusion module is designed, a cross-attention fusion feature pyramid network and a corresponding aircraft target detection model are constructed, and the information flow and fusion strategy between features are optimized by fully integrating features of different scales, and the dependency relationship between features of different scales is mined, thereby enhancing the ability to capture key features of the target; the model can be enabled to accurately identify and locate aircraft targets in complex backgrounds, significantly improving the accuracy and robustness of detection, so as to realize accurate aircraft target detection task recognition. The specific implementation methods of the present invention are as follows:

[0045] 1. Build a cross-attention fusion module:

[0046] In an embodiment of the present invention, a Cross Attention Fusion Module (CAFM) performs attention calculation on feature maps of adjacent scales to obtain feature dependencies between different scales. Specifically, the Cross Attention Fusion Module takes feature maps of adjacent scales as input, and the feature maps of two different scales are upsampled or downsampled before being input into the Cross Attention Fusion Module to unify the sizes of the two inputs. The specific upsampling and downsampling implementation strategies will be described in detail in step 2.

[0047] First, the two feature maps are transformed into embedding vectors that can be calculated by the attention mechanism through deformation operations. Suppose the input feature maps are F a With F b , then the deformation operation can be expressed as:

[0048]

[0049] Among them, ReshapeA is a deformation operation that converts the three-dimensional feature map into an embedded vector. and are the transformed embedding vectors respectively, and then, through a layer normalization operation, that is, and Get the input vector X of the multi-head attention module Ea With X Eb .

[0050] Combination Figure 2 and Figure 3 As shown in the figure, in the multi-head attention mechanism, each "head" performs attention calculations independently, and multiple heads can be processed in parallel. In this object detection model, 8 attention heads are set. Through this parallel computing method, the model can extract input features from multiple angles and integrate these diverse perspective information to generate a richer and more comprehensive attention feature representation.

[0051] In X Ea With X Eb Before being sent to the multi-head attention, each will be copied twice, that is, three identical X Ea With X Eb , and make an exchange, that is, two X Ea Vector with an X Eb The vector will be sent to the same multi-head attention module for cross attention calculation. At the same time, the two X Eb Vector with an X Ea The vector will be sent to another multi-head attention module for cross-attention calculation. In this embodiment, each attention head has its own three independent fully connected layers, mapping the three input vectors into query (Query, Q), key (Key, K) and value (Value, V), where the query comes from the exchanged input vector. The specific operation can be expressed as the following formula:

[0052]

[0053] in, and They represent the independent learnable weight matrices in the jth attention head. In the attention mechanism, the dot product of the query vector and the key vector is first calculated and scaled to obtain the attention score. Subsequently, these scores are normalized and the normalized attention scores are multiplied by the corresponding value vector to complete the subsequent calculation process. Both multi-head attentions perform the above calculation operations. The attention calculation process can be formally expressed as follows:

[0054]

[0055] in, and Represents the output of each head of two multi-head attention. k Represents the dimension of the key vector, which is used for scaling operations to prevent excessively large values ​​from being generated before applying the softmax function, thereby ensuring the stability of numerical calculations. The softmax function is a normalized exponential function that is used to convert attention scores into probability distributions, reflecting the degree of attention each output element pays to each element in the input feature. According to the obtained probability distribution, the value vector is weighted and summed to form the final output of the attention mechanism. This process can be understood as extracting and combining important features from the input features according to the attention weights. Finally, the outputs of all attention heads are concatenated and processed through a linear transformation layer to integrate the information extracted by each head. Both multi-head attentions perform the above calculation operations, which are formally expressed as follows:

[0056]

[0057] Among them, W O is a learnable weight matrix, representing the linear transformation layer processing, and Concat represents the concatenation operation.

[0058] Finally, the multi-head attention output is sequentially subjected to residual connection and layer normalization operations, then passed through a multi-layer perceptron (MLP), and finally subjected to a residual connection operation and layer normalization operation. The whole process can be formally described as follows:

[0059]

[0060] AttentionOut a =LayerNorm(Output a +MLP(Output a ))

[0061] Another cross-attention branch performs the same corresponding operation to obtain AttentionOut b , and then AttentionOut a With AttentionOut b Perform a concatenation operation, and convert the concatenated embedding vector back to a three-dimensional feature map through a deformation operation. The specific operation can be expressed as the following formula:

[0062] CAO=ReshapeB(Concat(AttentionOut a ,AttentionOut b ))

[0063] Among them, CAO (Cross Attention Output) is the final output of the cross attention, and ReshapeB is the reverse deformation operation of ReshapeA. The complete cross attention fusion calculation can be expressed by the following formula:

[0064] CAO=CAFM(F a ,F b )

[0065] Among them, CAFM represents the computation of the complete cross-attention fusion module.

[0066] 2. Build a cross-attention fusion feature pyramid network:

[0067] In an embodiment of the present invention, the Cross Attention Fusion Feature Pyramid Network (CAF-FPN) takes a multi-scale feature map as input. Figure 4 As shown, let the input multi-scale feature map be C i In this target detection task, there are four feature maps of different scales, denoted as c1, C2, C3, and C4. Among them, the scale of C3 is twice that of C4, the scale of C2 is twice that of C3, and the scale of C1 is twice that of C2. These four initial feature maps need to be initially fused. Specifically, the dimensions of feature maps of different scales are adjusted through 1×1 convolution operations to obtain feature maps F1, F2, F3, and F4. This operation can be formally expressed as follows:

[0068] F i =Conv 1×1 (C i )

[0069] Furthermore, the obtained feature maps F1, F2, F3, and F4 will be fused between adjacent feature maps through the cross attention fusion module. First, F3 is downsampled to make its scale consistent with F4, and then F4 and the downsampled F3 are input into the cross attention fusion module for calculation. Then, the attention calculation output is passed through a convolution layer to obtain the fused feature map P4. The specific operation can be expressed by the following formula:

[0070] CAO4=CAFM(F4,Downsampling(F3))

[0071] P4=Conv 3×3 (CAO4)

[0072] Among them, Downsampling represents the downsampling operation, CAO4 represents the intermediate fusion feature obtained, and the scale of CAO4 is consistent with F4. Similarly, CAO4 is upsampled to make its scale consistent with F3, and then the upsampled CAO4 and F3 are input into the cross-attention fusion module for calculation. In addition, the obtained cross-attention fusion output feature is then cross-attentive fused with the downsampled F2 again. The specific operation can be expressed as follows:

[0073]

[0074] P3=Conv 3×3 (CAO3)

[0075] Among them, Upsampling represents the upsampling operation. and CAO3 represent the intermediate fusion features. Therefore, P3 includes the dependency relationship with the features of the previous layer and the next layer through the cross attention fusion module. The calculation method of the fusion feature P2 is consistent with that of P3, as shown in the following formula:

[0076]

[0077] P2=Conv 3×3 (CAO2)

[0078] To obtain P1, CAO2 is upsampled and then cross-attention fusion is performed with F1, and then P1 is obtained through a convolution layer, as shown in the following formula:

[0079] CAO1=CAFM(F1,Downsampling(CAO2))

[0080] P1=Conv 3×3 (CAO1)

[0081] Finally, the obtained multi-scale fusion feature maps include P1, P2, P3, and P4. The calculation of the entire cross-attention fusion pyramid network can be summarized as:

[0082] P1,P2,P3,P4=CAFFPN(C1,C2,C3,C4)

[0083] Among them, CAFFPN stands for Cross Attention Fusion Pyramid Network.

[0084] 3. Build a complete aircraft target detection model:

[0085] like Figure 5As shown in the figure, the complete aircraft target detection model mainly consists of image preprocessing, feature extraction network, cross-attention fusion feature pyramid network, and detection head.

[0086] First, remote sensing images need to go through image preprocessing, which includes a series of image enhancement operations, such as horizontal flipping, noise injection, image sharpening and blurring, etc. Then, the preprocessed image is sent to the multi-scale feature extraction network for multi-scale feature extraction, which can be specifically expressed as the following formula:

[0087] C1,C2,C3,C4=BackboneNet(I)

[0088] Among them, I is the remote sensing image after image preprocessing, BackboneNet is a multi-scale feature extraction network, and a common network can be selected, such as the ResNet network.

[0089] Subsequently, the obtained multi-scale feature maps C1, C2, C3, and C4 will be used as the input of the cross-attention fusion feature pyramid network, in which the cross-attention feature fusion calculation is performed to obtain the fused multi-scale feature maps P1, P2, P3, and P4.

[0090] Finally, the detection head receives the fused multi-scale feature map and accurately locates and classifies the aircraft target. Specifically, the detection head uses the regression network to regress the anchor boxes in the feature maps of each level based on the predefined anchor boxes to accurately determine the position of the aircraft detection box. At the same time, the classification network performs classification operations on all feature points in the feature map to evaluate the category probability of each detection box. In order to optimize the detection results, the low-confidence detection boxes are first removed by confidence threshold screening, and then the non-maximum suppression (NMS) technology is applied to eliminate redundant detection boxes. Combining these two steps, the final precise positioning and classification results are output, thereby effectively achieving the target detection task. The specific operation can be expressed as follows:

[0091] y=Score(ClsNet(P1,P2,P3,P4))

[0092] bbox=NMS(RegNet(P1,P2,P3,P4))

[0093] Among them, RegNet represents the regression network and ClsNet represents the classification network, both of which are composed of multi-layer convolutional networks. The classification network contains a Sigmoid activation function in its last layer, which is used to convert the output into a probability value between 0 and 1. Score represents the confidence filtering mechanism, which is used to discard the detection results with confidence lower than the set threshold; NMS is responsible for filtering overlapping prediction boxes to retain the best detection results. Let the detection head be DetHead, then the complete prediction calculation process can be formally expressed as follows:

[0094] y,bbox=DetHead(P1,P2,P3,P4)

[0095] Among them, y corresponds to bbox one by one, bbox represents the predicted detection box, and y represents the predicted probability of the category to which the detection box belongs.

[0096] 4. Training and testing the model on remote sensing image aircraft data:

[0097] In the embodiment of the present invention, after constructing an aircraft target detection model based on a cross-attention fusion feature pyramid network, the model uses a cross-entropy loss function (Cross-entropy Loss, L CE ) and Mean Squared Error loss function (Mean Squared Error, L MSE ) for training. Finally, the total loss function of the model is the sum of these two loss functions, which is specifically expressed as:

[0098] L CE =L CE (y,y label )+L MSE (bbox,bbox label )

[0099] Among them, bbox label With y label Represent the true label values ​​of the detection box and classification respectively. When the value of the loss function tends to be stable and no longer decreases significantly during the model training process, it indicates that the model has converged and reached a stable state. At this time, the training process ends, and the aircraft target detection model based on the cross-attention fusion feature pyramid network is obtained.

[0100] Then, the trained aircraft target detection model based on the cross-attention fusion feature pyramid network is used to perform target detection experiments on the test images. The process can be formally expressed as follows:

[0101]

[0102] in, (Cross-Attention-Fusion-Feature-Pyramidbased DetectionModel) represents the aircraft target detection model based on the cross-attention fusion feature pyramid network obtained through training; I Test represents the remote sensing test image to be detected; and y test and bbox test They respectively represent the category probability and positioning box of the aircraft target detection results generated during the test.

[0103] From the description of the above embodiments, those skilled in the art can know that the present invention provides an aircraft target detection method based on a cross-attention fusion feature pyramid network, wherein a cross-attention fusion module is constructed, and a cross-attention fusion feature pyramid network and its corresponding aircraft target detection model are constructed in combination with the cross-attention fusion module, and the aircraft target detection model is used to perform remote sensing image aircraft target detection without manual intervention, and accurate remote sensing image aircraft target detection can be achieved; the aircraft target detection model based on the cross-attention fusion feature pyramid network proposed by the method of the present invention enriches the semantic information of the fusion feature and improves the multi-scale feature expression ability by effectively utilizing the advantages of the attention mechanism in capturing the association of global semantic information. This feature enables the model to more comprehensively extract and characterize data features in the remote sensing image aircraft target detection task, better understand the image content, and thus achieve high-precision aircraft target detection. The method of the present invention provides a more reliable and efficient solution for the field of aircraft target detection, and provides a new perspective and possibility for the development of aircraft target detection technology.

[0104] Furthermore, the present invention also provides an aircraft target detection system based on a cross-attention fusion feature pyramid network, which is applied to an aircraft target detection method based on a cross-attention fusion feature pyramid network in the above embodiment to improve the performance of remote sensing image target detection and facilitate accurate aircraft target detection task recognition; the system includes:

[0105] A target detection model building module is used to build a cross-attention fusion module, build a cross-attention fusion feature pyramid network in combination with the cross-attention fusion module, and build an aircraft target detection model based on the cross-attention fusion feature pyramid network;

[0106] A target detection model training module, used for training the aircraft target detection model for an aircraft target detection task;

[0107] The target detection model application module is used to use the trained aircraft target detection model to perform aircraft target detection on the remote sensing images to be tested.

[0108] The implementation principle and technical effects of the system provided in the embodiment of the present invention are the same as those of the aforementioned method embodiment. For the sake of brief description, for matters not mentioned in the system embodiment, reference can be made to the corresponding contents in the aforementioned method embodiment, which will not be repeated here.

[0109] In addition, an embodiment of the present invention also provides a storage medium on which one or more programs readable by a computing device are stored, and the one or more programs include instructions. When the instructions are executed by the computing device, the computing device executes an aircraft target detection method based on a cross-attention fusion feature pyramid network in the above-mentioned embodiment.

[0110] In the embodiment of the present invention, the storage medium may be, for example, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples of storage media (a non-exhaustive list) include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, and any suitable combination thereof.

[0111] Those skilled in the art should understand that the embodiments of the present invention can be provided as methods, systems or computer program products, etc. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects.

[0112] It should be noted that the word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention can be implemented by means of hardware comprising several distinct components, and by means of a suitably programmed computer.

[0113] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0114] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for detecting aircraft targets based on a cross-attention fusion feature pyramid network, characterized in that: The method comprises the following steps: S1, build cross attention fusion module; S2. Build a cross-attention fusion feature pyramid network in combination with the cross-attention fusion module; S3. Building an aircraft target detection model based on the cross-attention fusion feature pyramid network; S4, training the aircraft target detection model; and performing aircraft target detection on the remote sensing image to be measured using the trained aircraft target detection model.

2. The method for detecting aircraft targets based on a cross-attention fusion feature pyramid network according to claim 1, characterized in that: In step S1, the cross-attention fusion module constructed includes: a layer normalization module, a multi-head attention module and a multi-layer perceptron module; the cross-attention fusion module takes the feature maps of adjacent scales as input, performs attention calculation on the feature maps of adjacent scales, and obtains the feature dependency between different scales.

3. The method for detecting aircraft targets based on a cross-attention fusion feature pyramid network according to claim 2, characterized in that: In the multi-head attention module, each attention head performs attention calculation independently, and multiple attention heads are processed in parallel.

4. The method for detecting aircraft targets based on a cross-attention fusion feature pyramid network according to claim 3 is characterized in that: In the step S2, in the constructed cross-attention fusion feature pyramid network, a multi-scale feature map is used as input, and the dimensions of feature maps of different scales are adjusted through convolution operations. The adjusted feature maps are subjected to feature fusion between adjacent feature maps through a cross-attention fusion module, and then a multi-scale fused feature map is obtained through a convolution layer.

5. The method for detecting aircraft targets based on a cross-attention fusion feature pyramid network according to claim 4, characterized in that: In step S3, the aircraft target detection model constructed includes: an image preprocessing network, a feature extraction network, a cross-attention fusion feature pyramid network and a detection head connected in sequence, wherein: The image preprocessing network performs image preprocessing operations on remote sensing images; the preprocessed image data is sent to the feature extraction network for multi-scale feature extraction; the obtained multi-scale features are sent to the cross-attention fusion feature pyramid network to perform cross-attention feature fusion calculations to obtain a multi-scale fusion feature map; the multi-scale fusion feature map is input into the detection head to locate and classify aircraft targets.

6. The method for detecting aircraft targets based on a cross-attention fusion feature pyramid network according to claim 5, characterized in that: The image preprocessing operations of the image preprocessing network include: horizontal flipping, noise injection, image sharpening and blurring; The detection head includes a regression network and a classification network. The detection head uses the regression network to regress the anchor frames in the feature maps of each level based on the predefined anchor frames to accurately determine the position of the aircraft detection frame; at the same time, the classification network performs classification operations on all feature points in the feature map to evaluate the category probability of each detection frame.

7. The method for detecting aircraft targets based on a cross-attention fusion feature pyramid network according to claim 1, characterized in that: In step S4, the model is trained for the remote sensing image aircraft target detection task using a cross entropy loss function and a mean square error loss function.

8. An aircraft target detection system based on a cross-attention fusion feature pyramid network, characterized in that: The method for detecting an aircraft target based on a cross-attention fusion feature pyramid network as described in any one of claims 1 to 7 is applied to realize accurate aircraft target detection task recognition, and the system comprises: A target detection model building module is used to build a cross-attention fusion module, build a cross-attention fusion feature pyramid network in combination with the cross-attention fusion module, and build an aircraft target detection model based on the cross-attention fusion feature pyramid network; A target detection model training module, used for training the aircraft target detection model for an aircraft target detection task; The target detection model application module is used to use the trained aircraft target detection model to perform aircraft target detection on the remote sensing images to be tested.

Images