Deep learning-based air weak and small target detection method

Abstract

The invention discloses an air weak and small target detection method based on deep learning. The method comprises the following steps: acquiring a to-be-detected air image; inputting the to-be-detected aerial image into a target detection model to obtain an initial detection result; wherein the target detection model is obtained through training of a training set, the training set comprises an extreme environment aerial image and a label, the target detection model is constructed by adopting a neural network, and the neural network is used for performing multiple image fusion on a to-be-detected aerial image to extract a confidence map; and obtaining a detection result according to the confidence map. According to the method, a feature embedding module is introduced into an air weak and small target detection model to assist detection, so that an excellent detection effect can be shown under the conditions of a complex background and a low signal-to-noise ratio.

Description

Technical Field

[0001] The present invention belongs to the technical field of target detection, and in particular relates to a method for detecting small aerial targets based on deep learning. Background Art

[0002] The performance of aerial target detection technology in complex backgrounds and low-contrast conditions is particularly important, especially in the fields of drone monitoring, aerial reconnaissance, and emergency response. Low contrast and environmental interference often lead to blurring and loss of target features, thus affecting the accuracy of detection and recognition. For example, in drone monitoring, environmental noise may hinder the detection of potential targets; in aerial reconnaissance, the reduced effectiveness of the detection system under low-light conditions may affect the success of the mission. Therefore, improving the robustness of aerial target detection technology is of great significance to improving the accuracy of these tasks in harsh environments.

[0003] For the detection of small targets in aerial images, existing methods mainly include algorithms based on physical models and deep learning. Methods based on physical models, such as image enhancement and denoising, improve target detection by increasing the overall contrast of aerial images or the brightness of specific areas, but their performance is limited under low signal-to-noise ratio conditions. Deep learning methods such as YOLO and Faster R-CNN learn the detection and recognition features of targets by training neural networks on large-scale datasets. These methods show good detection performance under complex backgrounds, but still have certain limitations under extreme environments, such as high noise and low contrast conditions. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a method for detecting small aerial targets based on deep learning, which can show good detection performance in extreme environments.

[0005] The present invention provides a method for detecting small and weak aerial targets based on deep learning, comprising:

[0006] Acquire the aerial image to be detected;

[0007] The aerial image to be detected is input into the target detection model to obtain an initial detection result; wherein, the target detection model is obtained by training with a training set, and the training set includes: aerial images of extreme environments and annotations, and the target detection model is constructed using a neural network, and the neural network is used to perform multiple image fusions on the aerial image to be detected to extract a confidence map, and obtain the detection result based on the confidence map.

[0008] Optionally, the target detection model includes: a feature embedding module, a position encoding module, a composite encoding module, an upsampling module and a decoder;

[0009] The feature embedding module is used to transform the low-level feature representation of the input image into a high-dimensional feature representation;

[0010] The position encoding module is used to add spatial position information to the input sequence;

[0011] The composite coding module is used to extract local and global features and provide multi-scale semantic information;

[0012] The upsampling module is used to obtain a high-resolution feature map;

[0013] The decoder is used to decode the high-resolution feature map to obtain an initial detection result.

[0014] Optionally, the feature embedding module includes: a convolutional layer, an activation layer and a plurality of convolutional blocks connected in sequence;

[0015] The convolution block includes: a convolution layer, a batch normalization layer and an activation layer connected in sequence.

[0016] Optionally, the position encoding module includes: a feature embedding layer and an encoding layer;

[0017] The feature embedding layer is used to segment and linearly embed the input image to generate an embedded feature sequence;

[0018] The coding layer is used to introduce spatial information to perform position coding on the feature sequence to obtain a sequence feature graph.

[0019] Optionally, the composite encoding module includes: a multi-head self-attention unit and an enhancement unit;

[0020] The multi-head self-attention unit is used to obtain global context information according to the sequence feature graph;

[0021] The enhancement unit is used to enhance feature expression by aggregating context information at different positions and generate an enhanced feature map with the same dimension as the sequence feature map.

[0022] Optionally, the multi-head self-attention unit includes: a self-attention mechanism layer and a global normalization layer;

[0023] The enhancement unit includes: a first convolutional layer and a second convolutional layer.

[0024] Optionally, the method is characterized in that obtaining global context information includes:

[0025] Input the sequence feature graph into the self-attention mechanism layer in sequence to obtain a first result graph;

[0026] Inputting the sequence feature graph into a global normalization layer to obtain a second result graph;

[0027] The first result graph and the second result graph are fused to obtain the global context information.

[0028] Optionally, generating an enhanced feature map having the same dimension as the sequence feature map includes:

[0029] Inputting the global context information into the first convolutional layer and the second convolutional layer in sequence to obtain a third result graph;

[0030] The third result map and the global context information are weightedly fused to obtain the enhanced feature map.

[0031] Optionally, decoding the high-resolution feature map to obtain a detection result includes:

[0032] The high-resolution feature map is fused with the sequence feature map, and the fused feature map is decoded to obtain the detection result.

[0033] Optionally, the loss function is:

[0034]

[0035] Among them, L iou is the loss function, y is the confidence map, x gt is a real labeled image.

[0036] Compared with the prior art, the present invention has the following advantages and technical effects:

[0037] The method of the present invention introduces a feature embedding module into the aerial small target detection model to assist detection, so that excellent detection effects can be achieved even under complex backgrounds and low signal-to-noise ratio conditions. The model uses aerial image features as prior information related to the characteristics of small targets, and introduces a feature extraction strategy that combines a Transformer module with a composite encoder module. The Transformer module can extract global context information of the target area, which complements the local feature information of the composite encoder module to further improve the detection performance. In addition, a multi-head self-attention mechanism and a layer normalization module are designed in the context enhancement module of the detection model to better identify key feature channels and improve the detection accuracy of the model. The context enhancement module can effectively focus on important areas in the image and reduce attention to irrelevant backgrounds, so that the model can achieve efficient small target detection in different environments, ultimately improving the detection performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0039] Figure 1 This is a flow chart of a method for detecting small aerial targets based on deep learning according to an embodiment of the present invention;

[0040] Figure 2 is a schematic diagram of the structure of image detection according to an embodiment of the present invention;

[0041] Figure 3 is a schematic diagram of a local structure of a target detection model according to an embodiment of the present invention;

[0042] Figure 4 is a schematic diagram of a Transformer module according to an embodiment of the present invention;

[0043] Figure 5 This is a diagram showing the effect of detecting a small and weak target in the air according to an embodiment of the present invention. DETAILED DESCRIPTION

[0044] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0045] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0046] The present invention proposes a method for detecting small aerial targets based on deep learning. Figure 1 As shown, the specific steps include:

[0047] Acquire the aerial image to be detected;

[0048] The aerial image to be detected is input into the target detection model to obtain the initial detection result; wherein, the target detection model is obtained by training with a training set, the training set includes aerial images of extreme environments and annotations, and the target detection model is constructed using a neural network, and the neural network is used to perform multiple image fusions on the aerial image to be detected to extract a confidence map, and obtain the detection result based on the confidence map.

[0049] Furthermore, the target detection model includes: a feature embedding module, a position encoding module, a composite encoding module, an upsampling module and a decoder;

[0050] Feature embedding module, used to transform the low-level feature representation of the input image into a high-dimensional feature representation;

[0051] Position encoding module, used to add spatial position information to the input sequence;

[0052] Composite encoding module, used to extract local and global features and provide multi-scale semantic information;

[0053] Upsampling module, used to obtain high-resolution feature maps;

[0054] The decoder is used to decode the high-resolution feature map and obtain the initial detection results.

[0055] Specifically, Figure 2 As shown, obtaining an image to be detected;

[0056] The aerial image is input into the trained object detection model, in which the aerial image is passed through a feature embedding module to obtain a first feature map;

[0057] The first feature map is subjected to block segmentation and linear embedding to generate an embedded feature sequence, and position encoding is performed to obtain a second feature map;

[0058] The second feature map is input into the composite encoder module, which includes a preset number of feature extraction structures connected in sequence. When passing through the feature extraction structure, the input of each feature extraction structure passes through the multi-head self-attention module (MSA), the layer normalization module, the context enhancement module (CAM) and the first activation layer in sequence to obtain a third feature map, and then the third feature map is fused with the input of the feature extraction structure to obtain a fourth feature map; the fourth feature map then passes through the second multi-head self-attention module, the second layer normalization module, the context enhancement module (CAM) and the second activation layer in the feature extraction structure in sequence to generate a fifth feature map, and then the fifth feature map is fused with the fourth feature map to obtain a sixth feature map, and the sixth feature map is fused with the input of the feature extraction structure to obtain a seventh feature map as the output feature of the composite encoder module.

[0059] The seventh feature map is subjected to an upsampling module to obtain a high-resolution eighth feature map;

[0060] In the decoding module, the eighth feature map is jump-connected with the second feature map to obtain a ninth feature map;

[0061] The ninth feature map is decoded to generate a confidence map.

[0062] Finally, the confidence map is input into the output module to obtain the detection result map.

[0063] Furthermore, the feature embedding module includes: a convolutional layer, an activation layer, and a number of convolutional blocks connected in sequence;

[0064] The convolution block consists of: a convolutional layer, a batch normalization layer, and an activation layer connected in sequence.

[0065] Specifically, the feature embedding module includes a convolution layer, an activation layer, and several convolution blocks connected in sequence from data input to output, and each convolution block includes a convolution layer, a batch normalization layer, and an activation layer connected in sequence from data input to output.

[0066] Further, the position encoding module includes: a feature embedding layer and an encoding layer;

[0067] The feature embedding layer is used to segment and linearly embed the input image to generate an embedded feature sequence;

[0068] The encoding layer is used to introduce spatial information to perform position encoding on the feature sequence and obtain the sequence feature map.

[0069] Specifically, the encoding module includes a feature embedding layer, which performs block segmentation and linear embedding on the input aerial image to generate an embedded feature sequence, and introduces spatial information through position encoding to obtain a sequence feature map.

[0070] Furthermore, the composite encoding module includes: a multi-head self-attention unit and an enhancement unit;

[0071] Multi-head self-attention unit, used to obtain global context information based on sequence feature graph;

[0072] The enhancement unit is used to enhance the feature expression by aggregating contextual information at different positions and generate an enhanced feature map with the same dimension as the sequence feature map.

[0073] Specifically, the composite encoder module combines a multi-head self-attention module (MSA) with a context enhancement module (CAM), and performs feature extraction through multiple stages, each of which includes a multi-head self-attention layer and a context enhancement layer connected in sequence.

[0074] The multi-head self-attention module (MSA) in the composite encoder module is used to calculate the global correlation of input features and extract global context information; the context enhancement module (CAM) enhances feature expression by aggregating context information at different positions and generates an enhanced feature map with the same dimension as the input feature map.

[0075] Furthermore, the multi-head self-attention unit includes: a self-attention mechanism layer and a global normalization layer;

[0076] The enhancement unit includes: a first convolutional layer and a second convolutional layer.

[0077] Furthermore, it is characterized in that obtaining global context information includes:

[0078] Input the sequence feature graph into the self-attention mechanism layer in sequence to obtain the first result graph;

[0079] Inputting the sequence feature graph into the global normalization layer to obtain a second result graph;

[0080] The first result graph and the second result graph are fused to obtain global context information.

[0081] Specifically, the multi-head self-attention module includes a parallel self-attention mechanism and a global normalization module. The output of the second-layer normalization module is processed by the self-attention mechanism and the global normalization module respectively, and then the output of the self-attention mechanism is fused with the output of the global normalization module and input into the context enhancement module (CAM).

[0082] Furthermore, generating an enhanced feature map having the same dimension as the sequence feature map includes:

[0083] Inputting the global context information into the first convolutional layer and the second convolutional layer in sequence to obtain a third result graph;

[0084] The third result map and the global context information are weightedly fused to obtain an enhanced feature map.

[0085] Specifically, the context enhancement module includes a first convolutional layer and a second convolutional layer. The output of the second multi-head self-attention module is processed by the first convolutional layer and the second convolutional layer in sequence, and then fused with the output of the second multi-head self-attention module to obtain an intermediate feature map. At the same time, the output of the second multi-head self-attention module is processed by the second convolutional layer in sequence, fused with the intermediate feature map, and then input into the first activation layer.

[0086] Furthermore, the high-resolution feature map is decoded to obtain the detection results including:

[0087] The high-resolution feature map is fused with the sequence feature map, and the fused feature map is decoded to obtain the detection result.

[0088] Specifically, the decoding module includes a preset number of upsampling layers connected in sequence, each of which includes a transposed convolutional layer for restoring the high-dimensional feature map output by the composite encoder to the spatial resolution of the target detection result.

[0089] The upsampling module is used to restore the spatial resolution of the feature map and perform jump connections with the output features of each stage of the composite encoder, so that the low-level detail information is integrated with the high-level semantic information to improve the accuracy of target detection; the final output feature map is decoded to generate a confidence map and a detection result map.

[0090] Furthermore, the loss function is:

[0091]

[0092] Among them, L iouis the loss function, y is the confidence map, x gt is a real labeled image.

[0093] Specifically, the loss function in the training of the aerial small target detection model uses the intersection over union (IoU) to calculate the distance between the confidence map and the true annotation, which helps to deal with the category imbalance problem between the target and the background, and focuses more on the area of ​​the aerial small target. It is defined as follows:

[0094]

[0095] Among them, y is the confidence map, x gt is a real labeled image.

[0096] Furthermore, the preconditions are:

[0097] The initial detection result partially overlaps with the true annotation and the pixel distance between the center of the initial detection result and the center of the true annotation is less than a preset threshold, wherein the center of the detection result is obtained by calculating the centroid of the binary segmentation mask.

[0098] Specifically, the conditions for the detection result to be correct are that the following two points are met at the same time: (1) the output result overlaps with the true annotation in some pixels; (2) the pixel distance between the center of the detection result and the center of the true annotation is less than a certain threshold (4 pixels). Five widely used evaluation indicators are used, including detection probability (P d ), P d -F a The area under the curve (AUC), target level F1 value (F t 1) Pixel level (F p 1) and Intersection over Union (IoU), these indices are defined as follows:

[0099]

[0100] Target level F1 value F t 1 and pixel level F1 value F p 1 is defined as follows:

[0101]

[0102] The precision and recall are defined as:

[0103]

[0104] Among them, T p is a true example, F p is a false true example, F N is a false negative example. The Intersection over Union (IoU) is a pixel-level metric defined as:

[0105]

[0106] The present embodiment is described in detail below with reference to the accompanying drawings:

[0107] like Figure 1-Figure 5 As shown, in one embodiment, a method for detecting small aerial targets based on deep learning includes:

[0108] Acquire an image to be detected;

[0109] The image is input into the trained small object detection model, in which:

[0110] The aerial image is passed through the feature embedding module to obtain the first feature map, and then the embedded feature sequence is generated through block segmentation and linear embedding, and position encoding is performed to obtain the second feature map;

[0111] Among them, the feature embedding module includes a convolution layer, an activation layer and a plurality of convolution blocks connected in sequence from data input to output, and each convolution block includes a convolution layer, a batch normalization layer and an activation layer connected in sequence from data input to output.

[0112] It should be noted that after the aerial image is input into the target detection model, when passing through the feature embedding module, this part will increase the number of channels to extract the features of the aerial image; the size of the input aerial image is C×H×W (where C represents the number of channels, H and W represent the height and width of the image respectively, and the specific size is determined according to the actual input image), and it passes through the convolution layer (using 3×3 convolution layer, at this time the number of channels increases from 3 to 64), the activation layer (using softmax activation function, at this time the feature map size is 64×H×W) and multiple sequentially connected convolution blocks (at this time the first feature map is obtained, and the size of the first feature map is 64×H×W), wherein each convolution block includes a convolution layer (using 3×3 convolution layer), a batch normalization layer and an activation layer (using softmax activation function) sequentially connected from data input to output.

[0113] Among them, position encoding helps the model understand the relative or absolute position of each element in the input data when processing sequence data. Position encoding adds a position encoding vector to the input embedding vector to provide the position information of each element in the sequence. Specifically: for the posth position in the input sequence and the embedding dimension i:

[0114] When i is even:

[0115] When i is an odd number:

[0116] Calculate the encoding vector for each position to get a matrix with the same length as the input sequence, and add this matrix to the input embedding matrix element by element to get the final input representation.

[0117] The second feature map is input into the composite encoder module to obtain a third feature map;

[0118] Among them, Figure 4 As shown in the figure, the composite encoder module is based on the Transformer model and includes multiple feature extraction structures in sequence. Each feature extraction structure contains a multi-head self-attention module (MSA), a layer normalization module, a context enhancement module (CAM), and a first activation layer. After the second feature map (size 64xHxW) is input, it passes through the first feature extraction structure. After the feature map passes through the multi-head self-attention module, the number of channels increases to 128, while the height and width remain unchanged, and the size of the third feature map is 128xHxW. Next, after layer normalization and activation function processing, the feature map still maintains this size, but the eigenvalue distribution is optimized.

[0119] The third feature map is fused with the second feature map to obtain a fourth feature map;

[0120] During the fusion process, the third feature map is added to the input second feature map (size 64xHxW), and the size of the obtained fourth feature map remains 128xHxW, combining feature information from two different sources.

[0121] The fourth feature map is processed by the second feature extraction structure in turn to obtain a fifth feature map;

[0122] After the second feature extraction structure, the feature map passes through the multi-head self-attention module again, and the size becomes, at this time the height and width are halved, and the number of channels is doubled. Then, it is processed by layer normalization and the second activation layer to further enhance the feature expression ability.

[0123] The fifth feature map is merged with the fourth feature map to obtain a sixth feature map;

[0124] Among them, the fifth feature map after being processed by the second feature extraction structure has a size of Add the fourth feature map to obtain the sixth feature map size To further integrate feature information at different levels.

[0125] Then the sixth feature map is passed through the attention module, the residual module, the decoding module and the output module in sequence, and finally the seventh feature map is obtained. (The backbone module, the encoding module, the attention module, the residual module, the decoding module and the output module constitute the CNN network designed by the present invention).

[0126] The seventh feature map is input into the upsampling module, and an eighth feature map is obtained after upsampling processing;

[0127] The upsampling module uses a bilinear interpolation method to upsample the feature map to 256xHxW. At this point, the upsampling module expands the spatial dimension of the feature map by interpolating the pixels of each feature map, so as to effectively fuse features with the subsequent decoding module.

[0128] In the decoding module, the eighth feature map is jump-connected with the second feature map to obtain a ninth feature map;

[0129] Among them, the size of the second feature map is 64xHxW. By concatenating the feature with the eighth feature map (this concatenation is the concatenation of the channel dimension), the size of the ninth feature map is 320xHxW (256+64 channels), thereby enhancing the model's ability to recover detailed information. This jump connection ensures that during the decoding process, important low-level features can be passed to high-level structures, thereby improving the performance of the final output.

[0130] Subsequently, the ninth feature map passes through a series of convolutional layers, activation functions, and the final output layer in the decoding module to generate the final output feature map;

[0131] The convolution layer in the decoding module uses a 3x 3 convolution kernel to extract deep features and gradually restore spatial resolution through layer-by-layer convolution. After the ninth feature map passes through the first convolution layer, the feature map size is 320x H x W. After the activation layer (using the ReLU activation function), the nonlinearity of the feature map is maintained. Subsequently, after the second convolution layer and the activation layer, the feature map size is still 320x H x W.

[0132] Finally, after passing through the output layer (usually a 1x1 convolutional layer), the output feature map is obtained. The output feature map is as follows Figure 5 As shown in the figure, the size is NxHxW, where N is the number of categories output by the model. This process helps the model make full use of contextual information and effectively detect small and weak targets in the air, improving detection precision and accuracy.

[0133] The data set used for model training in this embodiment consists of 15,064 images of various flying objects, with a training rate of 80% and a verification rate of 20%. Each image is annotated with the class number of the object and the coordinates of the edge of the relevant bounding box. An image may have multiple objects and classes, with an average of 1.6 annotated objects per image and a total of 24,769 annotations for all images. The median image scale is 416x416. These images have been preprocessed for automatic orientation, but no enhancement has been applied. The data set represents a long-tail distribution, in which drones (25.2% of objects), birds (25%), airplanes (7.9%), and helicopters (6.3%) account for most of the data set (64.4%), and there is a class imbalance.

[0134] The framework of the method proposed in this embodiment is implemented using Pytorch 1.7.1 and accelerated by CUDA 11.2. The entire network is trained using the SGD algorithm with a learning rate of 0.01, a momentum of 0.9, and a weight decay of 1e-4, running on an NVIDIA GeForce RTX4050 GPU. The batch size is set to 24 and the maximum number of training rounds is 150. Before training, all input images are normalized. Subsequently, the images are data enhanced by random flipping. Finally, the images are resized to 224x224. The number of heads of MSA is set to 12, and the number of encoding layers is set to 12.

[0135] The loss function in the training of the small aerial target detection model uses the intersection over union (IoU) to calculate the distance between the confidence map and the true annotation, which helps to deal with the category imbalance problem between the target and the background, and focuses more on the area of ​​small aerial targets. It is defined as follows:

[0136]

[0137] Among them, y is the confidence map, x gt is a real labeled image.

[0138] The method for detecting small aerial targets based on deep learning is characterized in that the detection result is correct when the following two conditions are met at the same time: (1) the output result overlaps with the true annotation in some pixels; (2) the pixel distance between the center of the detection result and the center of the true annotation is less than a certain threshold (4 pixels). Five widely used evaluation indicators are used, including detection probability (P d ), P d -F a The area under the curve (AUC), target level F1 value (F t 1) Pixel level (F p 1) and Intersection over Union (IoU), these indices are defined as follows:

[0139]

[0140] Target level F1 value F t 1 and pixel level F1 value F p 1 is defined as follows:

[0141]

[0142] The precision and recall are defined as:

[0143]

[0144] Among them, T pis a true example, F p is a false true example, F N is a false negative example. The Intersection over Union (IoU) is a pixel-level metric defined as:

[0145]

[0146] The deep learning-based aerial small target detection method introduces a feature embedding module into the aerial small target detection model to assist detection, so that it can also show excellent detection results under complex backgrounds and low signal-to-noise ratio conditions. At the same time, the aerial image is regarded as prior information related to the target features, and the Transformer module is combined with the improved UNet network to extract features. Transformer can obtain global information including target features, which complements the local feature information extracted by UNet, thereby improving the detection performance; the target detection model helps the model to better focus on the key feature channels in the image through attention, thereby improving the detection accuracy of the model. The attention mechanism enables the model to focus on important areas in the image, reduce interference with the background, and enable the model to accurately detect small targets even in complex backgrounds, thereby improving the detection performance.

[0147] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for detecting small aerial targets based on deep learning, characterized in that: include: Acquire the aerial image to be detected; The aerial image to be detected is input into the target detection model to obtain an initial detection result; wherein, the target detection model is obtained by training with a training set, and the training set includes: aerial images of extreme environments and annotations, and the target detection model is constructed using a neural network, and the neural network is used to perform multiple image fusions on the aerial image to be detected to extract a confidence map, and obtain the detection result based on the confidence map.

2. The method for detecting small aerial targets based on deep learning according to claim 1, characterized in that: The target detection model includes: a feature embedding module, a position encoding module, a composite encoding module, an upsampling module and a decoder; The feature embedding module is used to transform the low-level feature representation of the input image into a high-dimensional feature representation; The position encoding module is used to add spatial position information to the input sequence; The composite coding module is used to extract local and global features and provide multi-scale semantic information; The upsampling module is used to obtain a high-resolution feature map; The decoder is used to decode the high-resolution feature map to obtain an initial detection result.

3. The method for detecting small aerial targets based on deep learning according to claim 2, characterized in that: The feature embedding module includes: a convolutional layer, an activation layer and a plurality of convolutional blocks connected in sequence; The convolution block includes: a convolution layer, a batch normalization layer and an activation layer connected in sequence.

4. The method for detecting small aerial targets based on deep learning according to claim 2, characterized in that: The position encoding module includes: a feature embedding layer and an encoding layer; The feature embedding layer is used to segment and linearly embed the input image to generate an embedded feature sequence; The coding layer is used to introduce spatial information to perform position coding on the feature sequence to obtain a sequence feature graph.

5. The method for detecting small aerial targets based on deep learning according to claim 4, characterized in that: The composite coding module includes: a multi-head self-attention unit and an enhancement unit; The multi-head self-attention unit is used to obtain global context information according to the sequence feature graph; The enhancement unit is used to enhance feature expression by aggregating context information at different positions and generate an enhanced feature map with the same dimension as the sequence feature map.

6. The method for detecting small aerial targets based on deep learning according to claim 5, characterized in that: The multi-head self-attention unit includes: a self-attention mechanism layer and a global normalization layer; The enhancement unit includes: a first convolutional layer and a second convolutional layer.

7. The method for detecting small aerial targets based on deep learning according to claim 6, characterized in that: Obtaining global context information includes: Input the sequence feature graph into the self-attention mechanism layer in sequence to obtain a first result graph; Inputting the sequence feature graph into a global normalization layer to obtain a second result graph; The first result graph and the second result graph are fused to obtain the global context information.

8. The method for detecting small aerial targets based on deep learning according to claim 6, characterized in that: Generating an enhanced feature map of the same dimension as the sequence feature map includes: Inputting the global context information into the first convolutional layer and the second convolutional layer in sequence to obtain a third result graph; The third result map and the global context information are weightedly fused to obtain the enhanced feature map.

9. The method for detecting small and weak aerial targets based on deep learning according to claim 4, characterized in that: Decoding the high-resolution feature map to obtain a detection result includes: The high-resolution feature map is fused with the sequence feature map, and the fused feature map is decoded to obtain the detection result.

10. The method for detecting small aerial targets based on deep learning according to claim 4, characterized in that: The loss function is used for convergence during the training process, and the loss function is: Among them, L iou is the loss function, y is the confidence map, x gt is a real labeled image.

Images