A deep learning-based backlight x-ray image multi-layer noise removal method

Abstract

The application provides a backlight X-ray image multilayer noise removing method based on deep learning, which comprises the following steps: collecting a spherical shell contour in a backlight X-ray image, and obtaining a noise-free image dataset by computer modeling; simulating image noise and adding the same to the noise-free image dataset to obtain a noisy image dataset; constructing an image noise reduction model with the FR-Unet architecture as the core, inputting the noisy image dataset, performing image enhancement, and outputting a noise-reduced noisy image; comparing the noise-reduced noisy image with the noise-free image to calculate a mean square error loss (MSE), and training multiple times to obtain an optimal image noise reduction model; and applying the optimal image noise reduction model to a real experimental image to verify the noise reduction effect of the model and ensure the image quality and detail retention. The application can effectively remove multilayer noise, realize efficient noise reduction while maintaining high image fidelity, avoid loss of important information, and improve the quality and accuracy of inertial confinement fusion experimental data analysis.

Description

Technical Field

[0001] The present invention belongs to the technical field of X-ray imaging noise reduction in inertial confinement fusion (ICF) experiments, and in particular relates to a method for removing multi-layer noise from backlit X-ray images based on deep learning. Background Art

[0002] Back-illuminated X-ray imaging plays a vital role in inertial confinement fusion (ICF) experiments. Traditionally, the method of using film packages to capture back-illuminated X-ray signals has significant time inefficiency and limited resolution. To address these problems, digital detectors such as charge-coupled devices (CCDs) and complementary metal-oxide-semiconductor (CMOS) cameras have been introduced, which can provide fast and high-resolution imaging data. However, these advanced digital detectors also bring new challenges, namely that they are susceptible to shot noise, periodic speckle noise, and electronic noise, which originate from radiation dose, the scintillator used, and the imaging system itself. The presence of these noises seriously hinders the accurate observation and analysis of the internal evolution processes of matter in real experiments.

[0003] Currently, X-ray image noise reduction technologies are mainly divided into two categories: traditional methods and deep learning-based methods. Traditional methods include spatial domain filtering (such as median filtering and mean filtering), frequency domain filtering (such as Wiener filtering and Fourier transform), and various statistical methods (such as non-local means denoising). These methods can remove noise to a certain extent, but may cause edge blurring or loss of detail.

[0004] Deep learning-based methods, particularly convolutional neural networks (CNNs), have made significant progress in image processing in recent years. By learning from large amounts of annotated data, deep learning models can automatically learn and extract features from images, achieving more accurate noise reduction while preserving image details. This approach has the advantage of being able to handle more complex noise patterns, but it requires significant computing resources and training data. Summary of the Invention

[0005] In light of these challenges, this paper proposes a deep learning-based multi-layer noise removal method for backlit X-ray images to address the challenges posed by complex noise in backlit X-ray images within an ICF experimental environment. This method employs a frequency residual U-Net (FR-UNet) model to simulate the formation of different types of physical noise and, combined with an image synthesis algorithm to generate training data, effectively removes multi-layer noise, including shot noise, periodic speckle noise, and electronic noise.

[0006] Based on the above technical objectives, the present invention provides the following technical solutions:

[0007] A method for removing multi-layer noise from backlit X-ray images based on deep learning, which specifically includes:

[0008] S1. Design an image acquisition module to capture the spherical shell outline in the backlit X-ray image during the inertial confinement fusion experiment (ICF) using a digital CCD detector. Calibrate the optical and physical parameters using the spherical shell outline, and use a computer program to obtain the projection image of the X-ray-illuminated spherical shell as a noise-free image dataset.

[0009] S2. Design a noise simulation module to simulate image noise distribution similar to that in real experimental scenarios, and add the simulated image noise to the noise-free image dataset to obtain a noisy image dataset;

[0010] S3. Build an image denoising model with the FR-Unet architecture as the core, use the noisy image dataset obtained in step S2 as input, perform image enhancement, and output the denoised noisy image;

[0011] S4, calculating the loss function of the image denoising model; calculating the mean square error (MSE) of the denoised noisy image and the noise-free image dataset obtained in step S1 to evaluate the denoising effect of the model and train the model; obtaining the trained optimal image denoising model through multiple iterations and optimization;

[0012] S5. Apply and verify: Apply the optimal image denoising model to real inertial confinement fusion experimental images to verify the denoising effect of the model and ensure image quality and detail preservation.

[0013] Furthermore, step S2 specifically includes:

[0014] S21, intercepting the non-spherical shell projection part of the backlit X-ray image as the experimental image noise, and converting it into a double-precision floating point type;

[0015] S22. Design a cyclic sampling noise image block, traverse the experimental image noise from the center of the image with a fixed pixel interval in row and column steps to extract the image block; first perform detrending processing on each image block, calibrate the experimental image noise, and then calculate the radial average power spectrum of the experimental image noise;

[0016] S23, radially rotating the radial average power spectrum obtained in step S22 to obtain an average noise power spectrum, then performing an inverse Fourier transform to obtain a spatial convolution kernel of the average noise power spectrum, and performing spatial convolution with Gaussian white noise to obtain simulated shot noise;

[0017] S24, converting the experimental image noise from the spatial domain to the frequency domain through Fourier transform. The periodic structure in the acquired image appears as discrete and sharp peaks in the frequency domain. The corresponding frequency components are separated to obtain simulated periodic speckle noise.

[0018] S25 , adding the simulated shot noise obtained in step S23 and the simulated periodic speckle noise obtained in step S24 to the noise-free image dataset, thereby obtaining a noisy image dataset.

[0019] More specifically, in step S22: the detrending process includes: removing abnormally bright image areas through wavelet transform, and subtracting the mean value of the image block to obtain a zero-mean image block; the calculation of the radial mean power spectrum of the experimental image noise specifically includes: first performing a two-dimensional discrete Fourier transform on the calibrated experimental image noise to obtain a noise power spectrum, then performing a radial analysis on the noise power spectrum to obtain a radial NPS, and accumulating the radial NPS of each image block and then calculating the average to obtain the radial mean power spectrum of the experimental image noise.

[0020] Furthermore, step S3 specifically includes:

[0021] S31, preprocessing the input noisy image data set, adjusting the image pixel values ​​to between [-1, 1] by normalization;

[0022] S32, constructing a FR-Unet network model, wherein the FR-Unet network model includes three parts: a backbone feature extraction part, an enhanced feature extraction part, and an image reconstruction part;

[0023] S33. Input the normalized noisy image dataset into the FR-Unet network for multi-layer denoising to obtain a denoised noisy image.

[0024] Furthermore, the specific structure of the three parts of the FR-Unet network model is as follows:

[0025] Backbone feature extraction part: It includes n backbone feature extraction layers, each of which includes a self-learning frequency domain module, a dense residual module, and a maximum pooling layer. The backbone feature extraction module is used to extract and stack image features to obtain a series of preliminary effective image features.

[0026] Enhanced feature extraction part: includes m enhanced feature extraction layers, each of which includes a self-learning frequency domain module, a dense residual module and an upsampling layer; it is used to fuse preliminary effective image features and enhance the expressiveness of image features;

[0027] Image reconstruction part: Through a convolution layer, the output features of the enhanced feature extraction part are resized to reconstruct a noise-free X-ray spherical shell projection image.

[0028] More specifically, the self-learning frequency domain module is used to achieve noise reduction and image enhancement. The specific implementation process includes:

[0029] In the self-learning frequency domain module, Fourier transform is first performed to transform the image spatial features input to the self-learning frequency domain module into image frequency domain features;

[0030] The image frequency domain features are then multiplied by a complex convolution kernel to obtain a complex weight matrix that matches the dimension of the image frequency domain features. The real and imaginary parts of the complex weights in the matrix are recorded as trainable parameters R1 and R2, respectively. These two parameters are iteratively updated through backpropagation during the training process.

[0031] Multiply the frequency spectrum of the input image features by the complex weight matrix element by element to complete the linear transformation, thereby enhancing the low-frequency signal and weakening the high-frequency signal;

[0032] Finally, an inverse Fourier transform is performed to convert the image features from the frequency domain back to the spatial domain to obtain the image spatial features after image enhancement.

[0033] Furthermore, step S4 specifically includes:

[0034] S41. Define the mean square error (MSE) between the noise-reduced image I and the noise-free image K of size m*n as:

[0035]

[0036] Among them, I(i,j) and K(i,j) represent the grayscale value of the image at (i,j);

[0037] S42. Drive the process through the Adam optimization algorithm, and adaptively adjust the learning rate according to the mean square error (MSE) to enable the model to dynamically explore and converge to a set of optimal network parameters θ; through multiple cycles of iteration and optimization, each iteration aims to refine the network weights and gradually improve the model's ability to recover clear details from the noisy image after denoising, and finally output the trained optimal image denoising model.

[0038] Based on the above technical solution, the present invention has the following beneficial effects:

[0039] The method proposed in the present invention enhances the key information in the image while improving the nonlinear expression ability and alleviating the gradient vanishing problem by designing a denoising model that combines frequency domain processing with densely connected residual learning. It greatly improves the quality and accuracy of inertial confinement fusion experimental data analysis and promotes the development of nuclear fusion research. Compared with traditional methods, the solution provided by the present invention can achieve more efficient noise reduction while maintaining high image fidelity, avoiding the loss of important information. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 Schematic diagram of the backlit X-ray imaging system and the resulting diagram;

[0041] Figure 2 Schematic diagram for modeling backlight projection on a spherical shell;

[0042] Figure 3 Obtain flow charts for noise-free and noisy image datasets;

[0043] Figure 4 This is a simulation process diagram of experimental image noise;

[0044] Figure 5 Schematic diagram of the FR-Unet network model structure;

[0045] Figure 6 Comparison chart of noise reduction effect and pixel value curve;

[0046] Figure 7 The flowchart of the method proposed in the present invention is shown in FIG. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0048] Although the steps in the present invention are arranged with numbers, they are not intended to limit the order of the steps. Unless the order of the steps is clearly stated or the execution of a step requires other steps as a basis, the relative order of the steps can be adjusted. It is understood that the term "and / or" used herein refers to and covers any and all possible combinations of one or more of the associated listed items.

[0049] like Figure 7 As shown, the present invention proposes a method for removing multi-layer noise from backlit X-ray images based on deep learning, which specifically includes the following steps:

[0050] S1. Design an image acquisition module to capture the spherical shell outline in the backlit X-ray image during the inertial confinement fusion experiment (ICF) using a digital detector (CCD).

[0051] In this embodiment, Figure 1 As shown: Figure 1(a) describes a laser beam with a duration of 2.5 nanoseconds and an energy of 10 kJ irradiating a metal bar made of molybdenum material. The metal bar is 1.5 mm by 1.5 mm and 5 microns thick. The X-rays emitted from the back of the metal bar will be as follows Figure 1 The geometry of the hollow spherical shell shown in (b) is projected onto the detector. The hollow sphere is approximately 1.5 mm away from the backlight source, while the detector is placed approximately 15 mm away from the backlight source, resulting in a magnification effect of approximately 10 times. The hollow spherical shell is filled with deuterium gas with a density of 2.0±0.4 mg / cc; the spherical shell is made of CH plastic with a density of 1.05 g / cc. On the detector side, a scintillator is used to convert X-rays into visible light, which is then coupled and recorded by a frame camera equipped with a charge-coupled device (CCD);

[0052] Then the optical and physical parameters are calibrated by the spherical shell profile, such as Figure 2 As shown, the projection image of the X-ray irradiated spherical shell is obtained by computer program modeling as a noise-free image dataset;

[0053] S2. Design a noise simulation module to simulate image noise distribution similar to that in real experimental scenarios, and add the simulated image noise to the noise-free image dataset to obtain a noisy image dataset;

[0054] In order to effectively train a deep learning denoising model, a sufficiently large image dataset is required; however, in an experimental environment, obtaining a large number of noisy image resources is limited. Therefore, in order to solve this problem, the present invention generates more training data by simulating the noise in the experiment; and as a preferred embodiment, Figure 3 、 4 As shown, step S2 specifically includes:

[0055] S21, intercepting the non-spherical shell projection part of the backlit X-ray image as the experimental image noise, and converting it into a double-precision floating point type;

[0056] S22. Design a cyclic sampling noise image block, traverse the experimental image noise from the center of the image with a fixed pixel interval in row and column steps to extract the image block; first perform detrending processing on each image block, calibrate the experimental image noise, and then calculate the radial average power spectrum of the experimental image noise;

[0057] The detrending process includes: removing abnormally bright image areas by wavelet transform, and subtracting the mean value of the image block to obtain a zero-mean image block;

[0058] The radial average power spectrum of the experimental image noise is calculated specifically as follows: first, a two-dimensional discrete Fourier transform is performed on the calibrated experimental image noise to obtain a noise power spectrum, then a radial analysis is performed on the noise power spectrum to obtain a radial NPS, and the radial NPS of each image block is accumulated and then averaged to obtain a radial average power spectrum of the experimental image noise;

[0059] S23, radially rotating the radial average power spectrum obtained in step S22 to obtain an average noise power spectrum, and then performing inverse Fourier transform to obtain a spatial convolution kernel of the average noise power spectrum, and performing spatial convolution with Gaussian white noise to obtain simulated shot noise; as described in step 1, there is also periodic dark speckle noise in the imaging system; therefore, periodic speckle noise can be further added to the noisy image. Since there is granular noise in the experimental image, it is a periodic speckle noise. We add periodic speckle noise based on the experimental image. The purpose of doing so is to make the simulated image noise closer to the experimental noise, so that the network model can better remove the noise in the experiment;

[0060] S24. Convert the experimental image noise from the spatial domain to the frequency domain through Fourier transform. The periodic structure in the captured image appears as discrete and sharp peaks in the frequency domain. Separate the corresponding frequency components to obtain simulated periodic speckle noise.

[0061] S25, adding the simulated shot noise obtained in step S23 and the simulated periodic speckle noise obtained in step S24 to the noise-free image dataset, thus obtaining a noisy image dataset. Figure 4 A dataset of backlit spherical shell projection images from China.

[0062] In this example, supervised training of the FR-Unet network model requires paired training datasets to constrain and guide model learning. Each pair consists of a noisy image and its corresponding noise-free image. Therefore, a paired dataset was constructed for network model training: one consisting of the original, clear X-ray backlit spherical shell projection image (noise-free), and the other a processed image with simulated noise added (noisy). Using these paired datasets, the network model can be effectively trained to distinguish and remove noise, restoring image clarity.

[0063] S3. Build an image denoising model with the FR-Unet architecture as the core, use the noisy image dataset obtained in step S2 as input, perform image enhancement, and output the denoised noisy image;

[0064] As a preferred embodiment, Figure 5 As shown, step S3 specifically includes:

[0065] S31, preprocessing the input noisy image data set, adjusting the image pixel values ​​to between [-1, 1] by normalization;

[0066] S32, constructing a FR-Unet network model, wherein the FR-Unet network model includes three parts: a backbone feature extraction part, an enhanced feature extraction part, and an image reconstruction part;

[0067] The backbone feature extraction part includes n backbone feature extraction layers, each layer includes a self-learning frequency domain module, a dense residual module and a maximum pooling layer; the backbone feature extraction module is used to extract and stack image features to obtain a series of preliminary effective image features; Figure 5 As shown, modules 1-3 are the main feature extraction parts. Let the input image feature be F0, whose size is (H0, W0, C0). After module 1, the image feature F1 of size (H1, W1, C1) is obtained. H, W, C represent the image feature height, width and channel dimension respectively. The formula is expressed as:

[0068] F1=DownSample(LFT(DCR(F0)));

[0069] Among them, DCR() represents the dense residual module, LFT() represents the self-learning frequency domain module; DownSample() represents downsampling; similarly, image features F2 and F3 are obtained in sequence; module 4 serves as the connection between the backbone feature extraction part and the enhanced feature extraction part, and its output image feature F4 has the same height and width as F3, and only the channel dimension is different;

[0070] In this embodiment, the self-learning frequency domain module is used to achieve noise reduction and image enhancement. The specific implementation process includes:

[0071] In the self-learning frequency domain module, Fourier transform is first performed to transform the image spatial features input to the self-learning frequency domain module into image frequency domain features;

[0072] The image frequency domain features are then multiplied by a complex convolution kernel to obtain a complex weight matrix that matches the dimension of the image frequency domain features. The real and imaginary parts of the complex weights in the matrix are recorded as trainable parameters R1 and R2, respectively. These two parameters are iteratively updated through backpropagation during the training process.

[0073] Multiply the frequency spectrum of the input image features by the complex weight matrix element by element to complete the linear transformation, thereby enhancing the low-frequency signal and weakening the high-frequency signal;

[0074] Finally, an inverse Fourier transform is performed to convert the image features from the frequency domain back to the spatial domain to obtain the image spatial features after image enhancement;

[0075] Enhanced feature extraction part: includes m enhanced feature extraction layers, each layer includes a self-learning frequency domain module, a dense residual module and an upsampling layer; used to fuse preliminary effective image features and enhance the expression ability of image features; such as Figure 5 As shown in the figure, modules 5-7 are the enhanced feature extraction parts. Taking module 5 as an example, the image feature F4 is first upsampled and concatenated with the image feature F3 on the channel dimension C4. Then, it passes through the dense residual module and the self-learning frequency domain module in sequence to obtain the image feature F5 of size (H5, W5, C5). The formula is expressed as:

[0076] F5=DCR(LFT(Contact(F3,UpSample(F4))));

[0077] Among them, Contact() represents splicing, and UpSample() represents upsampling; similarly, image features F6 and F7 are obtained in sequence;

[0078] Image reconstruction: The output features of the enhanced feature extraction part are used to reconstruct the noise-free X-ray spherical shell projection image. That is, by performing a convolution operation on the output feature F7 of module 7, the image feature dimension is converted to (H0, W0, C0) to be consistent with the input image feature dimension, thereby reconstructing the X-ray backlit spherical shell projection image after noise reduction processing.

[0079] S33. According to the above steps, the normalized noisy image dataset is input into the FR-Unet network for multi-layer denoising, thereby obtaining a denoised noisy image.

[0080] S4, calculating the loss function of the image denoising model; calculating the mean square error (MSE) of the denoised noisy image and the noise-free image dataset obtained in step S1 to evaluate the denoising effect of the model and train the model; obtaining the trained optimal image denoising model through multiple iterations and optimization;

[0081] As a preferred embodiment, step S4 specifically includes:

[0082] S41. Define the mean square error (MSE) between the noise-reduced image I and the noise-free image K of size m*n as:

[0083]

[0084] Among them, I(i,j) and K(i,j) represent the grayscale value of the image at (i,j);

[0085] S42, driving the process by the Adam optimization algorithm, adaptively adjusting the learning rate according to the mean square error (MSE), so that the model dynamically explores and converges to a set of optimal network parameters θ; through multiple cycles of iteration and optimization, each iteration aims to refine the network weights, gradually improve the model's ability to recover clear details from the denoised noisy image, and finally output the trained optimal image denoising model.

[0086] S5, application and verification; apply the optimal image denoising model to real inertial confinement fusion experiment images, verify the denoising effect of the model, and ensure the image quality and detail retention. The final denoising effect, image quality and detail retention are shown in Figure 6

[0087] The first and third columns in the figure respectively represent the experimental data image and the image after denoising processing by the FR-Unet network, and the second column represents the effect comparison in the local frame. It can be seen that the method proposed in the application has good denoising effect.

[0088] The fourth column shows the pixel value distribution curve of the actual X-ray backlight image and the denoised X-ray backlight image in the central region of the image, and compares the pixel changes in the region of interest (ROI) of the original image and the denoised image. Among them, the red curve represents the pixel value distribution of the actual X-ray image, showing large fluctuations, which means that there is a significant difference between adjacent pixel values. On the contrary, the blue curve represents the pixel value distribution after denoising using the FR-UNet model proposed in the application, which is smoother, indicating that the difference between adjacent pixel values is smaller.

[0089] In summary, the frequency residual U-Net (FR-UNet) model is proposed in the application, which can effectively remove multi-layer noise including shot noise, periodic speckle noise and electronic noise by simulating the formation process of different types of physical noise and combining image synthesis algorithm to generate training data. Compared with traditional methods, the scheme provided by the application can achieve more efficient noise reduction while maintaining high image fidelity, avoiding loss of important information.

[0090] It is obvious to those skilled in the art that the application is not limited to the details of the above exemplary embodiments, and can be implemented in other specific forms without departing from the spirit or essential characteristics of the application. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting, and the scope of the application is defined by the appended claims rather than the above description, and therefore all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the application. Any reference signs in the claims should not be regarded as limiting the claims involved.

[0091] ​In addition, it should be understood that although this specification is described in terms of implementation methods, not every implementation method contains only one independent technical solution. This narrative method of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

Claims

1. A method for removing multi-layer noise from backlit X-ray images based on deep learning, characterized in that: The specific steps include: S1. Design an image acquisition module to capture the spherical shell contour in the backlit X-ray image during the inertial confinement fusion experiment (ICF) using a digital CCD detector. Calibrate the optical and physical parameters using the spherical shell contour. Use a computer program to model the calibrated parameters to obtain the X-ray spherical shell projection image as a noise-free image dataset. S2. Design a noise simulation module to simulate an image noise distribution similar to that in a real experimental scene, and add the simulated image noise to the noise-free image dataset to obtain a noisy image dataset; the simulated image noise includes simulated shot noise and simulated periodic speckle noise; S3. Build an image denoising model with the FR-Unet architecture as the core, take the noisy image dataset obtained in step S2 as input, perform image enhancement, and output the denoised noisy image; the FR-Unet network model consists of three parts: backbone feature extraction part, enhanced feature extraction part, and image reconstruction part; Enhanced feature extraction includes m Enhanced feature extraction layers, each of which includes a self-learning frequency domain module, a dense residual module, and an upsampling layer; It is used to fuse preliminary effective image features and enhance the expressiveness of image features. The specific implementation process includes: In the self-learning frequency domain module, Fourier transform is first performed to transform the image spatial features input to the self-learning frequency domain module into image frequency domain features; The image frequency domain features are then multiplied by a complex convolution kernel to obtain a complex weight matrix that matches the dimension of the image frequency domain features. The real and imaginary parts of the complex weights in the matrix are recorded as trainable parameters R1 and R2, respectively. These two parameters are iteratively updated through backpropagation during the training process. Multiply the frequency spectrum of the input image features by the complex weight matrix element by element to complete the linear transformation, thereby enhancing the low-frequency signal and weakening the high-frequency signal; Finally, an inverse Fourier transform is performed to convert the image features from the frequency domain back to the spatial domain to obtain the image spatial features after image enhancement; S4, calculating the loss function of the image denoising model; calculating the mean square error (MSE) of the denoised noisy image and the noise-free image dataset obtained in step S1 to evaluate the denoising effect of the model and train the model; obtaining the trained optimal image denoising model through multiple iterations and optimization; S5. Apply and verify: Apply the optimal image denoising model to real inertial confinement fusion experimental images to verify the denoising effect of the model and ensure image quality and detail preservation.

2. The method for removing multi-layer noise from backlit X-ray images based on deep learning according to claim 1, characterized in that: Step S2 specifically includes: S21, intercepting the non-spherical shell projection part of the backlit X-ray image as the experimental image noise, and converting it into a double-precision floating point type; S22. Design a cyclic sampling noise image block, traverse the experimental image noise from the center of the image with a fixed pixel interval in row and column steps to extract the image block; first perform detrending processing on each image block, calibrate the experimental image noise, and then calculate the radial average power spectrum of the experimental image noise; S23, radially rotating the radial average power spectrum obtained in step S22 to obtain an average noise power spectrum, then performing an inverse Fourier transform to obtain a spatial convolution kernel of the average noise power spectrum, and performing spatial convolution with Gaussian white noise to obtain simulated shot noise; S24, converting the experimental image noise from the spatial domain to the frequency domain through Fourier transform. The periodic structure in the acquired image appears as discrete and sharp peaks in the frequency domain. The corresponding frequency components are separated to obtain simulated periodic speckle noise. S25 , adding the simulated shot noise obtained in step S23 and the simulated periodic speckle noise obtained in step S24 to the noise-free image dataset, thereby obtaining a noisy image dataset.

3. The method for removing multi-layer noise from backlit X-ray images based on deep learning according to claim 2, characterized in that: In step S22: The detrending process includes: removing abnormally bright image areas by wavelet transform, and subtracting the mean value of the image block to obtain a zero-mean image block; The radial average power spectrum of the experimental image noise is calculated specifically as follows: first, a two-dimensional discrete Fourier transform is performed on the calibrated experimental image noise to obtain a noise power spectrum, then a radial analysis is performed on the noise power spectrum to obtain a radial NPS, and the radial NPS of each image block is accumulated and then averaged to obtain the radial average power spectrum of the experimental image noise.

4. The method for removing multi-layer noise from backlit X-ray images based on deep learning according to claim 1, characterized in that: Step S3 specifically includes: S31, preprocessing the input noisy image data set, adjusting the image pixel values ​​to between [-1, 1] by normalization; S32, build FR-Unet network model; S33. Input the normalized noisy image dataset into the FR-Unet network for multi-layer denoising to obtain a denoised noisy image.

5. The method for removing multi-layer noise from backlit X-ray images based on deep learning according to claim 4, characterized in that: The specific structure of the backbone feature extraction part and image reconstruction part of the FR-Unet network model is as follows: Backbone feature extraction part: including n backbone feature extraction layers, each of which includes a self-learning frequency domain module, a dense residual module, and a maximum pooling layer; The backbone feature extraction module is used to extract and stack image features to obtain a series of preliminary effective image features; Image reconstruction part: Through a convolution layer, the output features of the enhanced feature extraction part are resized to reconstruct a noise-free X-ray spherical shell projection image.

6. The method for removing multi-layer noise from backlit X-ray images based on deep learning according to claim 1, characterized in that: Step S4 specifically includes: S41. Define the mean square error (MSE) between the noise-reduced image I and the noise-free image K of size m*n as: ； Among them, I(i,j) and K(i,j) represent the grayscale value of the image at (i,j); S42. Drive the process through the Adam optimization algorithm, and adaptively adjust the learning rate according to the mean square error (MSE) to enable the model to dynamically explore and converge to a set of optimal network parameters θ; through multiple cycles of iteration and optimization, each iteration aims to refine the network weights and gradually improve the model's ability to recover clear details from the noisy image after denoising, and finally output the trained optimal image denoising model.

Images