Zero sample infrared super-resolution method based on fuzzy kernel generation

Abstract

The invention discloses a zero sample infrared super-resolution method based on generation of a fuzzy kernel, and relates to the technical field of super-resolution reconstruction, and the method comprises the following steps: data set preparation, degradation kernel estimation, super-resolution network construction, network model training, and reconstruction performance index calculation. According to the method, a new fuzzy kernel estimation mode is designed, a simple and efficient super-resolution network without pre-training is constructed, and experimental results show that the method shows excellent super-resolution effect and high robustness on a plurality of standard data sets; the method is suitable for super-resolution reconstruction of various low-resolution infrared images, and particularly has important application value in the fields with high image resolution requirements such as military monitoring, medical imaging and environment detection.

Description

Technical Field

[0001] The present invention relates to the technical field of super-resolution reconstruction, and in particular to a zero-sample infrared super-resolution method based on generating a blur kernel. Background Art

[0002] In infrared imaging technology, image resolution directly affects the accuracy of subsequent analysis and decision-making. Single image super-resolution (SISR) has attracted attention due to its wide application in computer vision, especially in the field of infrared images. It aims to restore high-resolution (HR) images with rich details and textures from low-resolution (LR) images. In recent years, methods based on zero-shot learning (ZSL) have provided new ideas for super-resolution reconstruction technology. In the field of infrared image super-resolution reconstruction, the combination of zero-shot learning methods is of great significance in solving the scarcity of infrared image data, the simple degradation of existing super-resolution algorithms, and significantly improving the detail performance and resolution of infrared images.

[0003] The Chinese patent publication number is "CN117422620A", and its name is "Infrared image super-resolution reconstruction method for real scenes based on deep learning". This method establishes an infrared image degradation model and combines channel attention and densely connected super-resolution networks. Low-resolution images with different degradation conditions are used to improve the robustness of the network. Taking into account the scarcity of infrared image data, it is proposed to use complex degradation models to synthesize training data to solve the problem that existing deep learning-based super-resolution methods rely on a large amount of training data. However, the designed degradation model is single, and the generalization ability in various scenarios is not strong, and the model performance is significantly reduced without large-scale data pre-training. Summary of the invention

[0004] The technical solution of the present invention to solve the above technical problem is to provide a zero-sample infrared super-resolution method based on generating a blur kernel, comprising the following steps:

[0005] S1. Data preprocessing: randomly selecting infrared low-resolution images to be processed from an infrared image dataset, wherein the dataset supports image formats of multiple infrared imaging devices;

[0006] S2. Degradation kernel estimation: Establish a generative adversarial network model, input noise degradation into the generator, and output the degradation kernel through the generator. The degradation kernel is convolved with the high-resolution image and superimposed with Gaussian noise to generate a degraded low-resolution image I LR′ ;Crop a region in the generated image as a generated patch, and input it into the discriminator together with the real patch cropped from the input image for authenticity judgment. When the discriminator has difficulty distinguishing the authenticity, determine that the weight of the generator is close to the real degradation kernel;

[0007] S3. Constructing a super-resolution network: Use the designed derived convolution residual block to construct a super-resolution network and introduce two-way skip connection. The first way of skip connection adds the interpolated image and the output of the second convolution block. The main branch is used to predict the residual of the image. The second way of skip connection adds the output of the first convolution block and the output of the last derived convolution residual block.

[0008] S4. Training network model: Generate degradation kernel through adversarial training of generator and discriminator, convolve the generated degradation kernel with the image to be reconstructed and superimpose noise to generate degraded image, input it into super-resolution network for training, optimize loss function to improve performance, and generate reconstructed image;

[0009] S5. Calculation of reconstruction performance indicators: Peak signal-to-noise ratio and structural similarity index are used to objectively evaluate the quality of the reconstructed image.

[0010] Further, in step S2, the generative adversarial network model includes a generator and a discriminator;

[0011] The generator is used to simulate the degradation kernel of the input infrared image, generate a high-dimensional feature map through multiple convolution layers, and restore the degradation kernel of the original image through deconvolution operation. The generator contains 6 linear full convolution networks with a step size of 2. The convolution kernels of the first three layers are 7×7, 5×5, and 3×3, and the convolution kernels of the last three layers are 1×1. The last layer performs a blur degradation function with a step size of 2. Using random noise z as the initial input, the blur kernel is extracted through the convolution layer of the generator, and the output is used as the input of the next layer to extract the degradation kernel;

[0012] The discriminator is used to evaluate the authenticity of the degradation kernel output by the generator. A series of convolutional layers and pooling layers are used to reduce the dimension of the input image, and finally a binary classifier is used to determine whether the image is a real degradation kernel. The discriminator adopts a fully convolutional structure and uses a small receptive field of 7×7.

[0013] Further, in step S3, the derived convolution residual block includes two branches, a main branch consisting of 5×5 derived convolution, activation function and deconvolution operation, and a secondary branch consisting of residual connection;

[0014] The derived convolution is used to obtain multi-scale output features and greatly reduces the computational complexity and memory usage of the convolutional neural network for extracting quantitative features.

[0015] The deconvolution layer is interpolated by any interpolation operator, and then performs the same convolution operation as the convolution layer, so as to fully mine the information of the image or network features locally;

[0016] The residual connection is used to transfer the gradient of the deep layer directly to the shallow layer, bypassing the main branch, so that the shallow layer parameters can be effectively trained and the gradient vanishing problem can be effectively avoided.

[0017] Furthermore, in step S4, in the adversarial training of the generator and the discriminator, the loss function expression of the entire network is defined as follows:

[0018]

[0019] in, It is expressed as:

[0020]

[0021] Among them, α=0.5, β=0.5, γ=5, δ=1, is a regular term used to constrain the degenerate kernel extracted during the G network training process and is composed as follows:

[0022] Indicates that the cumulative sum of degenerate core elements is 1;

[0023] Indicates that m is a constant weighted mask, and the weight increases with the increase of the center distance. The closer to the boundary, the greater the penalty and the smaller the value;

[0024] Indicates the sparsity of the degenerate kernel and prevents over-smoothing;

[0025] It means that the center of mass of the degenerate kernel is the geometric center of the kernel.

[0026] Further, in step S4, a composite loss function is designed during training in the super-resolution network, including a pixel mean square error loss and a total variation regularization term;

[0027] Under the real condition degradation model, the loss function of the real condition degradation scenario is designed as follows:

[0028] loss2=MSE+λ2·loss TV ;

[0029] Among them, the MSE is the pixel mean square error loss, which is defined as follows:

[0030]

[0031] The loss TV is the total variation regularization term, defined as follows:

[0032] loss TV =∑ i,j ((x i,j-1 -x i,j ) 2 +(x i+1,j -xi,j ) 2 );

[0033] The Adam optimization algorithm is used in the training process, and the initial learning rate is selected as 0.001; the network is gradually optimized through an iterative process, and the function value threshold of the loss function is set at around 0.005.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] (1) The present invention provides an efficient degradation kernel estimation scheme, which uses a generative adversarial network to generate a blur kernel that can simulate a more realistic physical degradation process. The degradation kernel is adaptively adjusted according to the input image. This method obtains a degradation kernel instead of a network. This solves the problem that traditional methods are unable to capture complex blur patterns in blur kernel estimation. The resulting array can be directly applied to any super-resolution algorithm, has high portability, is suitable for different sensors and imaging systems, and has strong adaptability and versatility.

[0036] (2) The present invention uses the generated blur kernel to design a zero-sample infrared super-resolution method. This method does not require a large amount of data pre-training and can perform super-resolution reconstruction through an internal learning method using only a single low-resolution image. This solves the problem of lack of training data in specific application scenarios, improves the detail performance of super-resolution reconstruction, overcomes the previous limitation of relying on a large amount of labeled data, and significantly improves the reconstruction effect.

[0037] (3) The super-resolution variable rate reconstruction method proposed in this invention shows good results compared with the traditional super-resolution method. The experimental results show that in the experiment with a reconstruction factor of 2, the peak signal-to-noise ratio (PSNR) is improved by more than 0.9, and the structural similarity (SSIM) is improved by more than 0.03. In the experiments with reconstruction factors of 3 and 4, the evaluation indicators are also improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0039] Figure 1 A flowchart of the steps of the zero-sample infrared super-resolution method based on generating blur kernel according to the present invention;

[0040] Figure 2 It is a flow chart of the method for estimating infrared image degradation kernel of the present invention;

[0041] Figure 3 Generate a structural diagram of the network for the present invention;

[0042] Figure 4 It is a structural diagram of the discriminant network of the present invention;

[0043] Figure 5 It is a structural diagram of the super-resolution network of the present invention;

[0044] Figure 6 The structure diagram of the convolution residual block derived by the present invention;

[0045] Figure 7 This is a training block diagram of the super-resolution network of the present invention. DETAILED DESCRIPTION

[0046] The present invention proposes a zero-sample infrared super-resolution method based on generating blur kernel, aiming to design a zero-sample infrared super-resolution method based on generating blur kernel, which is used to solve the problem that the existing super-resolution methods based on deep learning rely on a large amount of training data.

[0047] The zero-sample infrared super-resolution method based on generating blur kernel proposed by the present invention is described below in a specific embodiment:

[0048] In the technical solution of this embodiment, Figure 1 As shown, a zero-sample infrared super-resolution method based on generating a blur kernel comprises the following steps:

[0049] S1. Data preprocessing: randomly selecting infrared low-resolution images to be processed from an infrared image dataset, wherein the dataset supports image formats of multiple infrared imaging devices;

[0050] Specifically, a random selection is made from the KAIST infrared image dataset to input the infrared low-resolution image to be processed. The input data of the generator is multiple 32×32 or 64×64 image blocks cropped from the infrared image. The input image will generate a blur kernel matching the image block through a convolution operation.

[0051] S2. Degradation kernel estimation: Establish a generative adversarial network model, such as Figure 2 As shown in Fig. 1, noise degradation is input into the generator, and the generator outputs the degradation kernel, which is convolved with the high-resolution image and superimposed with Gaussian noise to generate a degraded low-resolution image I LR′ ;Crop a region in the generated image as a generated patch, and input it into the discriminator together with the real patch cropped from the input image for authenticity judgment. When the discriminator has difficulty distinguishing the authenticity, determine that the weight of the generator is close to the real degradation kernel;

[0052] Specifically, a generation network (G) is established: Figure 3As shown in the figure, the generator is used to simulate the degradation kernel of the input infrared image, generate a high-dimensional feature map through multiple convolutional layers, and restore the degradation kernel of the original image through deconvolution operation.

[0053] The generator G contains 6 linear full convolution networks with a step size of 2. The convolution kernels of the first three layers are 7×7, 5×5, and 3×3, and the convolution kernels of the last three layers are 1×1. The last layer performs a blur degradation function with a step size of 2. The mathematical expression is:

[0054] D b (I)=(I LR *k)↓2;

[0055] Where * represents the convolution operation, k is the blur kernel of the degradation process, and the blurred degraded image can be obtained by convolving k with the input image;

[0056] Using random noise z as the initial input, each convolution kernel of the generator (G) is used as the convolution kernel of the convolution layer for extracting the blur kernel, and the output is used as the input of the next convolution layer for extracting the degradation kernel. The mathematical expression of the random noise vector z can be expressed as:

[0057] z=[z1,z2,…,z n ],z i ~P z ;

[0058] Where: z is a vector of dimension n, z i is the noise distribution P z The random variables P are independently sampled from z is a normal distribution, and n is the dimension of the noise vector.

[0059] P z is a normal distribution (Gaussian distribution), that is Each z i They are all sampled from the standard normal distribution, and the corresponding mathematical expression is:

[0060]

[0061] At this time, the input noise vector z of the generator is a standard normal distribution random vector with dimension n.

[0062] Then, the degraded kernel in the generator is convolved with the input high-resolution image, Gaussian noise is superimposed, and the random perturbation of Gaussian noise is used to interfere with the generated blur kernel to obtain a low-resolution image I LR′ , the formula is as follows:

[0063]

[0064] The degradation kernel extracted by the final generator is a set of small arrays that can be applied to the SR algorithm;

[0065] Establish a discriminant network (D), such as Figure 4 As shown in the figure, the discriminator is used to evaluate the authenticity of the degradation kernel output by the generator. A series of convolutional layers and pooling layers are used to reduce the dimension of the input image, and finally a binary classifier is used to determine whether the image is a real degradation kernel.

[0066] The discriminator uses a probability distribution map D similar to a heat map map To calculate the difference between each pixel, the real patch is judged as true, and the generated patch is judged as false, where D map Each value is the probability of each pixel being true. For a real patch, D map The output of should be close to 1; for the generated patch, D map The output should be close to 0.

[0067] A fully convolutional structure is adopted, with a small receptive field of 7×7, no pooling and stride, to ensure that D can judge each local patch separately and output the probability of each position for the discriminator to carefully distinguish the difference between the generated fake image patches and the real image patches.

[0068] S3. Constructing a super-resolution network: Use the designed derived convolution residual block to construct a super-resolution network and introduce two-way skip connection. The first way of skip connection adds the interpolated image and the output of the second convolution block. The main branch is used to predict the residual of the image. The second way of skip connection adds the output of the first convolution block and the output of the last derived convolution residual block.

[0069] Specifically, Figure 5 As shown in the figure, the super-resolution network uses four designed derived convolution residual blocks and introduces two-way jump connections. The first jump connection adds the interpolated image and the output of the second convolution block, and the main branch is used to predict the residual of the image; the second jump connection adds the output of the first convolution block and the output of the last derived convolution residual block;

[0070] like Figure 6 As shown in the figure, the derived convolution residual block is mainly divided into two branches: the main branch composed of 5×5 derived convolution filter, activation function and deconvolution operation, and the secondary branch composed of residual connection. The calculation process is as follows:

[0071] y=deconv(ReLU(conv(x)))+x;

[0072] Among them, ReLU, conv and deconv are activation functions, derived convolution and deconvolution operations respectively. From the input feature x to the output feature y, it is necessary to go through the operation of the main branch of deconv(ReLU(conv(x))) in the above formula, and the residual connection copies x directly backward to the deeper layers of the neural network. Through the secondary branch, the information of x is directly transmitted to the deeper layers of the neural network.

[0073] The derived convolution residual block can be integrated at any position in the network, and the "encoder-decoder" structure composed of derived convolution and deconvolution can be used to fully mine the information of image or network features locally.

[0074] S4. Training network model: Generate degradation kernel through adversarial training of generator and discriminator, convolve the generated degradation kernel with the image to be reconstructed and superimpose noise to generate degraded image, input it into super-resolution network for training, optimize loss function to improve performance, and generate reconstructed image;

[0075] Specifically, the generator and the discriminator perform adversarial training, the generator tries to generate results as close to the real degradation kernel as possible, while the discriminator tries to distinguish the generated image I LR′ Compared with the real image I LR When the loss of the generator and the discriminator reaches a balance, the weight of the generator will be used as the degradation kernel of the input of the super-resolution network to improve the subsequent super-resolution reconstruction effect.

[0076] The GAN was trained for 3000 iterations using the ADAM optimizer (β1 = 0.5; β2 = 0.999), alternating between single optimization steps of G and D, and setting the learning rate to 2e -4 , decaying × 0.1 every 750 times.

[0077] After the generator and discriminator training is completed, the generated blur kernel can be used in the subsequent zero-shot super-resolution reconstruction process.

[0078] S5. Calculation of reconstruction performance indicators: Peak signal-to-noise ratio and structural similarity index are used to objectively evaluate the quality of the reconstructed image.

[0079] Specifically, the corresponding evaluation indicators of the method proposed in the present invention are compared with the indicators of the prior art, such as Figure 5 As shown in the figure, it can be seen that the method proposed in the present invention has an increase in computational complexity and parameters compared with the existing method, but the structural similarity and peak signal-to-noise ratio are greatly improved; these indicators show that the present invention can achieve reconstruction of more accurate high-resolution images while ensuring the network operation speed.

[0080] PSNR and SSIM are used to objectively evaluate the quality of reconstructed images;

[0081] The peak signal-to-noise ratio can be simply defined by the mean square error, and its definition expression is shown as follows:

[0082]

[0083] Among them, l represents the number of grayscale bits of each pixel in the image, and the value is usually 8. At this time, the grayscale level of the pixels in the image is 256. The unit of peak signal-to-noise ratio is decibel (dB). The peak signal-to-noise ratio evaluation result has a strong similarity with the subjective evaluation result and is one of the most widely used objective quality evaluation indicators. From the definition, it can be seen that when the peak signal-to-noise ratio of the image to be evaluated is larger, the image distortion is smaller;

[0084] The structural similarity index is a comprehensive representation of the similarity in brightness, contrast, and correlation between the image to be evaluated and the reference image. Its definition is as shown in the formula:

[0085]

[0086] Among them, μ f Represents the grayscale average value of the reference image; Represents the grayscale average value of the image to be evaluated; σ f is the variance of the reference image; is the variance of the image to be evaluated; is the covariance between the grayscale values ​​of the reference image and the grayscale values ​​of the image to be evaluated; C1 and C2 are constants set to prevent the denominator from being zero.

[0087] Further, in step S2, the generative adversarial network model includes a generator and a discriminator;

[0088] The generator is used to simulate the degradation kernel of the input infrared image, generate a high-dimensional feature map through multiple convolution layers, and restore the degradation kernel of the original image through deconvolution operation. The generator contains 6 linear full convolution networks with a step size of 2. The convolution kernels of the first three layers are 7×7, 5×5, and 3×3, and the convolution kernels of the last three layers are 1×1. The last layer performs a blur degradation function with a step size of 2. Using random noise z as the initial input, the blur kernel is extracted through the convolution layer of the generator, and the output is used as the input of the next layer to extract the degradation kernel;

[0089] The discriminator is used to evaluate the authenticity of the degradation kernel output by the generator. A series of convolutional layers and pooling layers are used to reduce the dimension of the input image, and finally a binary classifier is used to determine whether the image is a real degradation kernel. The discriminator adopts a fully convolutional structure and uses a small receptive field of 7×7.

[0090] Further, in step S3, the derived convolution residual block includes two branches, a main branch consisting of 5×5 derived convolution, activation function and deconvolution operation, and a secondary branch consisting of residual connection;

[0091] The derived convolution is used to obtain multi-scale output features and greatly reduces the computational complexity and memory usage of the convolutional neural network for extracting quantitative features.

[0092] The deconvolution layer is interpolated by any interpolation operator, and then performs the same convolution operation as the convolution layer, so as to fully mine the information of the image or network features locally;

[0093] The residual connection is used to transfer the gradient of the deep layer directly to the shallow layer, bypassing the main branch, so that the shallow layer parameters can be effectively trained and the gradient vanishing problem can be effectively avoided.

[0094] Furthermore, in step S4, in the adversarial training of the generator and the discriminator, the loss function of the entire blur kernel estimation network is designed so that during the comparison process, the discriminator compares whether the local feature distributions at different scales are consistent, rather than the global size of the image. Therefore, even if the area extracted from the reduced image is relatively large, the discriminator can still learn the similarity between these feature distributions. The loss function expression of the entire network is defined as follows:

[0095]

[0096] in, It is expressed as:

[0097]

[0098] Among them, α=0.5, β=0.5, γ=5, δ=1, is a regular term used to constrain the degenerate kernel extracted during the G network training process and is composed as follows:

[0099] Indicates that the cumulative sum of degenerate core elements is 1;

[0100] Indicates that m is a constant weighted mask, and the weight increases with the increase of the center distance. The closer to the boundary, the greater the penalty and the smaller the value;

[0101] Indicates the sparsity of the degenerate kernel and prevents over-smoothing;

[0102] It means that the center of mass of the degenerate kernel is the geometric center of the kernel.

[0103] These items combined together help the model learn a centralized, sparse, and normalized convolution kernel for image processing or generation tasks. Ultimately, the generator continuously adjusts the downsampling kernel of its output image to keep the local features of the image at different scales consistent, thereby achieving optimization results.

[0104] Further, in step S4, training is performed in the super-resolution network, such as Figure 7 As shown in Figure 1, the super-resolution network designed by the present invention enriches the training set by performing four rotations (0°, 90°, 180°, 270°) and vertical and horizontal mirror reflections on each low-resolution (LR) and high-resolution (HR) image pair. To improve robustness and achieve a larger super-resolution scale factor (s) from a smaller LR image, the algorithm performs super-resolution processing step by step, applicable to multiple intermediate scale factors (s1, s2, ..., s m =s). At each intermediate scale factor si, the generated SR image HR i and their scaled / rotated versions are added to the progressively larger training set as new HR images. These images are scaled by a gradual factor s i+1 Scaling to generate new LR-HR training example pairs until the desired resolution increase is achieved.

[0105] A composite loss function is designed, including pixel mean square error loss and total variation regularization term; under the real condition degradation model, the loss function of the real condition degradation scenario is designed as follows:

[0106] loss2=MSE+λ2·loss TV ;

[0107] Among them, the MSE is the pixel mean square error loss, which is defined as follows:

[0108]

[0109] The loss TV is the total variation regularization term, defined as follows:

[0110] loss TV =∑ i,j ((x i,j-1 -x i,j ) 2 +(x i+1,j -x i,j ) 2 );

[0111] The Adam optimization algorithm is used in the training process, and the initial learning rate is selected as 0.001; the network is gradually optimized through an iterative process, and the function value threshold of the loss function is set at around 0.005.

[0112] When the value of the loss function is less than 0.005, it can be considered that the training of the entire network has been basically completed; after the training is completed, the parameters of the network are saved for future use. The super-resolution effect will be dynamically adjusted according to the current image quality during the training phase;

[0113] After training, the network will directly input the input low-resolution image into the optimized super-resolution network to generate a super-resolution output image.

[0114] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention 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 by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A zero-shot infrared super-resolution method based on generating blur kernels, characterized in that: The following steps are involved: S1. Data preprocessing: randomly selecting infrared low-resolution images to be processed from an infrared image dataset, wherein the dataset supports image formats of multiple infrared imaging devices; S2. Degradation kernel estimation: Establish a generative adversarial network model, input noise degradation into the generator, and output the degradation kernel through the generator. The degradation kernel is convolved with the high-resolution image and superimposed with Gaussian noise to generate a degraded low-resolution image I LR′ ;Crop a region in the generated image as a generated patch, and input it into the discriminator together with the real patch cropped from the input image for authenticity judgment. When the discriminator has difficulty distinguishing the authenticity, determine that the weight of the generator is close to the real degradation kernel; S3. Constructing a super-resolution network: Use the designed derived convolution residual block to construct a super-resolution network and introduce two-way skip connections. The first skip connection adds the interpolated image and the output of the second convolution block, and the main branch is used to predict the residual of the image. The second skip connection adds the output of the first convolutional block and the output of the last derived convolutional residual block; S4. Training network model: Generate degradation kernel through adversarial training of generator and discriminator, convolve the generated degradation kernel with the image to be reconstructed and superimpose noise to generate degraded image, input it into super-resolution network for training, optimize loss function to improve performance, and generate reconstructed image; S5. Calculation of reconstruction performance indicators: Peak signal-to-noise ratio and structural similarity index are used to objectively evaluate the quality of the reconstructed image.

2. The zero-sample infrared super-resolution method based on generating blur kernel according to claim 1, characterized in that: In step S2, the generative adversarial network model includes a generator and a discriminator; The generator is used to simulate the degradation kernel of the input infrared image, generate a high-dimensional feature map through multiple convolution layers, and restore the degradation kernel of the original image through deconvolution operation. The generator contains 6 linear full convolution networks with a step size of 2. The convolution kernels of the first three layers are 7×7, 5×5, and 3×3, and the convolution kernels of the last three layers are 1×1. The last layer performs a blur degradation function with a step size of 2. Using random noise z as the initial input, the blur kernel is extracted through the convolution layer of the generator, and the output is used as the input of the next layer to extract the degradation kernel; The discriminator is used to evaluate the authenticity of the degradation kernel output by the generator. A series of convolutional layers and pooling layers are used to reduce the dimension of the input image, and finally a binary classifier is used to determine whether the image is a real degradation kernel. The discriminator adopts a fully convolutional structure and uses a small receptive field of 7×7.

3. The zero-sample infrared super-resolution method based on generating blur kernel according to claim 1, characterized in that: In step S3, the derived convolution residual block includes two branches, a main branch consisting of 5×5 derived convolution, activation function and deconvolution operation, and a secondary branch consisting of residual connection; The derived convolution is used to obtain multi-scale output features and greatly reduces the computational complexity and memory usage of the convolutional neural network for extracting quantitative features. The deconvolution layer is interpolated by any interpolation operator, and then performs the same convolution operation as the convolution layer, so as to fully mine the information of the image or network features locally; The residual connection is used to transfer the gradient of the deep layer directly to the shallow layer, bypassing the main branch, so that the shallow layer parameters can be effectively trained and the gradient vanishing problem can be effectively avoided.

4. The zero-sample infrared super-resolution method based on generating blur kernel according to claim 1, characterized in that: In step S4, in the adversarial training of the generator and the discriminator, the loss function expression of the entire network is defined as follows: in, It is expressed as: Among them, α=0.5, β=0.5, γ=5, δ=1, is a regular term used to constrain the degenerate kernel extracted during the G network training process and is composed as follows: Indicates that the cumulative sum of degenerate core elements is 1; Indicates that m is a constant weighted mask, and the weight increases with the increase of the center distance. The closer to the boundary, the greater the penalty and the smaller the value; Indicates the sparsity of the degenerate kernel and prevents over-smoothing; It means that the center of mass of the degenerate kernel is the geometric center of the kernel.

5. The zero-sample infrared super-resolution method based on generating blur kernel according to claim 1, characterized in that: In step S4, a composite loss function is designed during training in the super-resolution network, including a pixel mean square error loss and a total variation regularization term; Under the real condition degradation model, the loss function of the real condition degradation scenario is designed as follows: loss2=MSE+λ2·loss TV ; Among them, the MSE is the pixel mean square error loss, which is defined as follows: The loss TV is the total variation regularization term, defined as follows: loss TV =∑ i,j ((x i,j-1 -x i,j ) 2 +(x i+1,j -x i,j ) 2 ); The Adam optimization algorithm is used in the training process, and the initial learning rate is selected as 0.001; the network is gradually optimized through an iterative process, and the function value threshold of the loss function is set at around 0.005.

Images