Small-angle scattering curve deblurring method and device based on machine learning

Through the machine learning of the fuzzy method of UNet network model, combined with the fuzzy and point diffusion function prediction model, the fuzzy problem of small angle scattered data under unknown instruments is solved, and more thorough fuzzy and point diffusion function prediction is achieved, improving data analysis accuracy.

CN119066351AActive Publication Date: 2024-12-03HUAZHONG UNIV OF SCI & TECH +1
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202411157961.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-22
Publication Date
2024-12-03
Estimated Expiration
2044-08-22

AI Technical Summary

Technical Problem

The prior art is difficult to completely eliminate the fuzzy effect of small angle scattered data in the case of unknown instrument point diffusion functions, affecting the data analysis process.

Method used

Using the UNet network model based on machine learning, combining the anti-fuzzy UNet network model and point diffusion function to predict the UNet network model, the small angle scattering curve after the fuzzy is generated through convolutional operations, and the point diffusion function is predicted in the process to achieve a more thorough Anti-fuzzing treatment.

Benefits of technology

In the case of unknown instrument point diffusion function, the fuzzy effect of small angle scattering data is effectively eliminated, the original characteristics of the small angle scattering curve are restored, and the prediction of the point diffusion function is realized, which improves the accuracy of data analysis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119066351A_ABST
    Figure CN119066351A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of small-angle scattering data processing, and particularly discloses a small-angle scattering curve deblurring method and device based on machine learning, and the method comprises the steps: obtaining a to-be-deblurred small-angle scattering curve, inputting the to-be-deblurred small-angle scattering curve to a pre-trained UNet weight model, and outputting the deblurred small-angle scattering curve; the UNet weight model is constructed based on a deblurring UNet network model and a point spread function prediction UNet network model; wherein the step of outputting the deblurred small-angle scattering curve comprises the following substeps of: inputting the small-angle scattering curve to be deblurred into a deblurring UNet network model to generate a deblurring result; inputting a one-dimensional array formed by the number of point spread function data points corresponding to the small angle scattering curve to be deblurred into a point spread function prediction UNet network model to generate a point spread function prediction result; and performing convolution operation based on the deblurring result and the point spread function prediction result to generate a small-angle scattering curve after deblurring.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application belongs to the technical field of small-angle scattering data processing, and more specifically, to a small-angle scattering curve deblurring method and device based on machine learning. Background Art

[0002] Small angle scattering refers to the scattering phenomenon of the sample under study within a very small angle near the incident light beam. It is caused by the uneven electron density at the nanoscale inside the sample. Small angle scattering technology has become an important means to study the submicroscopic structure and morphological characteristics of materials.

[0003] In small-angle scattering measurement, due to the wavelength broadening of the light source, the limited collimation of the collimation system and the limited spatial resolution of the detector, the measured small-angle scattering curve has a fuzzy effect with neither high peaks nor low valleys, which greatly affects the subsequent data analysis process. The main goal of deblurring is to eliminate the fuzzy effect caused by the point spread function (PSF) of the instrument and restore the original small-angle scattering curve characteristics.

[0004] Traditional deblurring methods are mainly divided into two categories: the first category is based on iterative estimation methods, which require the provision of an accurate point spread function and are an estimation iterative method, and have the problem of incomplete deblurring; the second category is based on direct algorithms based on deconvolution. Although this type of method only needs to provide an estimated point spread function, it also has the problem of incomplete deblurring.

[0005] The point spread function of an instrument usually needs to be obtained experimentally, which is relatively complicated. Therefore, how to deblur the small-angle scattering data under the premise of an unknown point spread function of the instrument and restore the original characteristics of the small-angle scattering curve is an urgent problem to be solved. Summary of the invention

[0006] In view of the defects existing in the related technology, the embodiments of the present application provide a small-angle scattering curve deblurring method and device based on machine learning, which aims to solve the problem of deblurring small-angle scattering data under the premise of unknown point spread function of the instrument.

[0007] In a first aspect, an embodiment of the present application provides a small-angle scattering curve deblurring method based on machine learning, comprising:

[0008] Obtaining a small-angle scattering curve to be deblurred;

[0009] The small-angle scattering curve to be deblurred is input into the pre-trained UNet weight model, and the deblurred small-angle scattering curve is output; the UNet weight model is constructed based on the deblurring UNet network model and the point spread function prediction UNet network model;

[0010] Among them, the output of the deblurred small-angle scattering curve includes:

[0011] The small-angle scattering curve to be deblurred is input into the deblurring UNet network model to generate a deblurring result; the one-dimensional array consisting of the number of point spread function data points corresponding to the small-angle scattering curve to be deblurred is input into the point spread function prediction UNet network model to generate a point spread function prediction result;

[0012] A convolution operation is performed based on the deblurring result and the point spread function prediction result to generate the deblurred small-angle scattering curve.

[0013] In a second aspect, the embodiment of the present application further provides a small-angle scattering curve deblurring device based on machine learning, comprising:

[0014] An acquisition unit, used for acquiring a small-angle scattering curve to be deblurred;

[0015] A deblurring unit is used to input the small-angle scattering curve to be deblurred into a pre-trained UNet weight model, and output the deblurred small-angle scattering curve; the UNet weight model is constructed based on the deblurring UNet network model and the point spread function prediction UNet network model;

[0016] The deblurring unit includes:

[0017] The deblurring UNet module is used to generate deblurring results based on the small-angle scattering curve to be deblurred;

[0018] The point spread function prediction UNet module is used to generate the point spread function prediction result based on the one-dimensional array consisting of the number of point spread function data points corresponding to the small-angle scattering curve to be deblurred;

[0019] The convolution module is used to perform convolution operation based on the deblurring result and the point spread function prediction result to generate a deblurred small-angle scattering curve.

[0020] In a third aspect, an embodiment of the present application further provides an electronic device, comprising: at least one memory for storing programs; and at least one processor for executing the programs stored in the memory. When the program stored in the memory is executed, the processor is used to execute the method described in the first aspect or any possible implementation of the first aspect.

[0021] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores a computer program. When the computer program runs on a processor, the processor executes the method described in the first aspect or any possible implementation of the first aspect.

[0022] In a fifth aspect, an embodiment of the present application further provides a computer program product, which, when executed on a processor, enables the processor to execute the method described in the first aspect or any possible implementation of the first aspect.

[0023] The small-angle scattering curve deblurring method based on machine learning provided in the embodiment of the present application utilizes the UNet network to realize the deblurring processing of the small-angle scattering curve under the premise of the point spread function of the unknown system; at the same time, the physical formation process of the small-angle scattering curve is fully considered in the deblurring process, thereby realizing a more thorough deblurring processing, and realizing the prediction of the point spread function while deblurring. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the present application or related technologies, the following is a brief introduction to the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0025] Figure 1 This is one of the flow charts of the small-angle scattering measurement data deblurring method based on machine learning provided in an embodiment of the present application;

[0026] Figure 2 This is the second flow chart of the small-angle scattering measurement data deblurring method based on machine learning provided in an embodiment of the present application;

[0027] Figure 3 It is a schematic diagram of the structure of three UNet networks connected in series provided in an embodiment of the present application;

[0028] Figure 4 This is one of the schematic diagrams of the small-angle scattering curve after deblurring provided in the embodiments of the present application;

[0029] Figure 5 It is one of the schematic diagrams of the point spread function prediction results provided in the embodiments of the present application;

[0030] Figure 6 This is the second schematic diagram of the small-angle scattering curve after deblurring provided in the embodiment of the present application;

[0031] Figure 7 This is the second schematic diagram of the point spread function prediction result provided in the embodiment of the present application;

[0032] Figure 8 It is a structural schematic diagram of a small-angle scattering measurement data deblurring device based on machine learning provided in an embodiment of the present application;

[0033] Fig. 9is a schematic diagram of the structure of a deblurring unit provided in an embodiment of the present application;

[0034] Fig.10 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0036] Figure 1 FIG. 1 is one of the flow charts of the small-angle scattering measurement data deblurring method based on machine learning provided in the embodiment of the present application, such as Figure 1 As shown, the method at least includes the following steps:

[0037] S101, obtaining a small-angle scattering curve to be deblurred.

[0038] S102, inputting the small-angle scattering curve to be deblurred into a pre-trained UNet weight model, and outputting the deblurred small-angle scattering curve.

[0039] Specifically, under the premise of an unknown point spread function of the instrument, consider using machine learning to deblur the small-angle scattering data and restore the original characteristics of the small-angle scattering curve. In the embodiment of the present application, UNet is specifically used as a deep learning network.

[0040] UNet is a convolutional neural network architecture for image segmentation, which is characterized by its U-shaped network structure. The UNet network structure consists of two parts: the encoder and the decoder. The encoder is responsible for gradually extracting the features of the input image and reducing the spatial resolution, while the decoder restores the features to the size of the original input image through upsampling operations and gradually generates segmentation results. UNet introduces skip connections in the decoder, which connects the feature map in the encoder with the corresponding feature map in the decoder. This skip connection can help the decoder better utilize feature information at different levels, thereby improving the accuracy of image segmentation and detail retention. Compared with other deep learning networks, the UNet network performs well in small sample situations and can be trained on less labeled data and achieve better results.

[0041] The small-angle scattering curve to be deblurred is input into the pre-trained UNet weight model, and the deblurred small-angle scattering curve is output. It should be noted that the small-angle scattering curve to be deblurred input into the UNet weight model and the deblurred small-angle scattering curve output both refer to the observed small-angle scattering curve, which is the result of taking the natural logarithm (ln) of the real or actual small-angle scattering curve.

[0042] The UNet weight model is constructed based on the defuzzification UNet network model and the point spread function prediction UNet network model. Accordingly, Figure 2 This is a second flow chart of a method for deblurring small-angle scattering measurement data based on machine learning provided in an embodiment of the present application, such as Figure 2 As shown, S102 specifically includes:

[0043] S201, inputting the small-angle scattering curve to be deblurred into the deblurring UNet network model to generate a deblurring result;

[0044] S202, inputting a one-dimensional array consisting of the number of point spread function data points corresponding to the small-angle scattering curve to be deblurred into a point spread function prediction UNet network model to generate a point spread function prediction result;

[0045] S203: Perform a convolution operation based on the deblurring result and the point spread function prediction result to generate a deblurred small-angle scattering curve.

[0046] Specifically, in S201, the deblurring UNet network model takes the small-angle scattering curve to be deblurred as input, and generates a deblurring result based on the small-angle scattering curve to be deblurred.

[0047] Optionally, the deblurring result output by the deblurring UNet model is the actual data after deblurring, that is, the true small-angle scattering curve, or, in order to facilitate the observation of the scattering characteristics, the actual data after deblurring is output and the natural logarithm is taken, that is, the deblurred small-angle scattering curve, also called the observed small-angle scattering curve.

[0048] In S202, the point spread function prediction UNet model takes as input a one-dimensional array consisting of the number of point spread function data points corresponding to the small-angle scattering curve to be deblurred, and generates a point spread function prediction result based on the one-dimensional array consisting of the number of point spread function data points corresponding to the small-angle scattering curve to be deblurred.

[0049] Optionally, the point spread function prediction result may be a Gaussian parameter or a point spread function corresponding to the point spread function. The Gaussian parameters corresponding to the point spread function include amplitude, mean (mu) and standard deviation (sigma), and the point spread function may be calculated by the Gaussian parameters, specifically satisfying the following calculation formula:

[0050]

[0051] It should be noted that the execution order of S201 and S202 can be replaced, or they can be executed simultaneously.

[0052] Optionally, the UNet weight model also outputs a point spread function prediction result.

[0053] In S203, based on the deblurring result output by the deblurring UNet network model and the point spread function prediction result output by the point spread function prediction UNet network model, a deblurred small-angle scattering curve is generated through a convolution operation.

[0054] Since the physical process of forming the blurred small-angle scattering curve is the convolution of the real small-angle scattering curve with the point spread function, and the actually observed small-angle scattering curve is the result after taking the natural logarithm (ln), S203 does not only perform a simple convolution operation. The convolution operation process in S203 is different for different situations of the deblurring result and the point spread function prediction result.

[0055] Optionally, if the deblurring result is a small-angle scattering curve after deblurring, and the point spread function prediction result is a point spread function, then S203 specifically includes:

[0056] The deblurred result is converted to the natural exponential (exp) and then convolved with the point spread function prediction result and the natural logarithm (ln) is taken to generate the deblurred small-angle scattering curve.

[0057] Optionally, if the deblurring result is a small-angle scattering curve after deblurring, and the point spread function prediction result is a Gaussian parameter corresponding to the point spread function, then S203 specifically includes:

[0058] The point spread function is determined based on the Gaussian parameters corresponding to the point spread function. The deblurred result is convolved with the point spread function after taking the natural exponential (exp) and taking the natural logarithm (ln) to generate the deblurred small-angle scattering curve.

[0059] Optionally, if the deblurring result is a real small-angle scattering curve after deblurring, and the point spread function prediction result is a point spread function, then S203 specifically includes:

[0060] The deblurred result is convolved with the point spread function prediction result and the natural logarithm (ln) is taken to generate the deblurred small-angle scattering curve.

[0061] Optionally, if the deblurring result is a real small-angle scattering curve after deblurring, and the point spread function prediction result is a Gaussian parameter corresponding to the point spread function, then S203 specifically includes:

[0062] The point spread function is determined based on the Gaussian parameters corresponding to the point spread function, and the deblurring result is convolved with the point spread function and the natural logarithm (ln) is taken to generate the deblurred small-angle scattering curve.

[0063] The small-angle scattering curve deblurring method based on machine learning provided in the embodiment of the present application utilizes the UNet network to realize the deblurring processing of the small-angle scattering curve under the premise of the point spread function of the unknown system; at the same time, the physical formation process of the small-angle scattering curve is fully considered in the deblurring process, thereby realizing a more thorough deblurring processing, and realizing the prediction of the point spread function while deblurring.

[0064] In some embodiments, the deblurring UNet network model and the point spread function prediction UNet network model are both composed of multiple UNet networks connected in series.

[0065] Specifically, in order to extract deeper curve features, improve the network's expressiveness and robustness, both the deblurring UNet network model and the point spread function prediction UNet network model need to have a deep network structure. Therefore, both the deblurring UNet network model and the point spread function prediction UNet network model are composed of multiple UNet networks connected in series.

[0066] For example, the deblurring UNet network model and the point spread function prediction UNet network model are both composed of three UNet networks connected in series. Figure 3 is a schematic diagram of the structure of three UNet networks connected in series provided in an embodiment of the present application, such as Figure 3 As shown in the figure, in each UNet network, the input passes through the downsampling area and the upsampling area in turn, and each layer in the downsampling area and the upsampling area is connected by a jump connection; in the downsampling area, the number of channels of the input is 64 after passing through the one-dimensional convolution, and then the number of channels is 128 after passing through the maximum pooling layer and the one-dimensional convolution, and finally the number of channels after passing through the maximum pooling layer and the one-dimensional convolution is 256; in the upsampling area, the number of channels of the input is 128 after passing through the one-dimensional convolution, and then the number of channels after passing through the maximum inverse pooling layer and the one-dimensional convolution is 64, and finally the number of channels after passing through the maximum inverse pooling layer and the one-dimensional convolution is 1.

[0067] In some embodiments, the activation function of the point spread function prediction UNet network model is a Softplus function.

[0068] Specifically, in order to prevent the occurrence of complex numbers and faster network convergence in the subsequent process of taking the natural exponential, the Softplus function is used as the point spread function to predict the activation function of the UNet network model.

[0069] Taking the prediction result of the point spread function as the Gaussian parameter corresponding to the point spread function as an example, the Softplus function is used to positively limit the output of the three values ​​of amplitude, mean and standard deviation. The specific form of the Softplus function is as follows:

[0070] f(x)=log(1+e x )

[0071] Among them, f(x) is the mathematical expression of the Softplus function, and x is the weighted input of the neuron.

[0072] In some embodiments, the UNet weight model is trained based on the following steps:

[0073] Obtaining small-angle scattering curves to be deblurred and after deblurring of different sample structures and different point spread functions as training data sets;

[0074] Based on the training data set, the deblurring UNet network model and the point spread function prediction UNet network model are trained respectively;

[0075] The trained UNet weight model is determined based on the trained deblurring UNet network model and the point spread function prediction UNet network model.

[0076] Specifically, different sample structures may include hard spheres, cylinders, ellipsoids, layered stacking structures, simple cubic crystals, face-centered cubic crystals, etc. Different point spread functions can be obtained by changing the standard deviation of the Gaussian function.

[0077] The small-angle scattering curves to be deblurred and deblurred of different sample structures and different point spread functions are obtained as training data sets. The training data sets are used to train the deblurring UNet network model and the point spread function prediction UNet network model respectively, so as to obtain the trained UNet weight model.

[0078] During the training of the deblurring UNet network model, the deblurring results output by the model are compared with the actual deblurred small-angle scattering curve, so as to achieve performance evaluation of the model training. During the training of the point spread function prediction UNet network model, the point spread function prediction results output by the model are compared with the actual point spread function, so as to achieve performance evaluation of the model training.

[0079] In some embodiments, the loss function of the UNet weight model adopts the mean squared error (MSE) loss function, which is specifically in the following form:

[0080]

[0081] Where N is the number of samples, y i is the predicted value of the model, Specifically, in some embodiments, two loss calculations are used, namely, the loss is calculated between the output curve of the deblurring module and the input label, and the loss is calculated between the output curve of the convolution module that generates the actual scattering curve and the input curve of the deblurring module.

[0082] Furthermore, during the training process, the optimizer selects the Adam optimizer, sets the learning rate to 0.0001, trains the samples in batches, calculates the loss function, and continuously updates the training parameters through the Adam optimizer. During the model training process, after each training cycle (Epoch), the accuracy of the validation set is counted. If the current accuracy is greater than the historical optimal accuracy, the trained label and weight model is saved.

[0083] The following is a specific example to further illustrate the training and application process of the UNet weight model proposed in the embodiment of the present application.

[0084] The small-angle scattering curves to be deblurred and after deblurring of different sample structures and different point spread functions are obtained as training data sets. The training data sets are divided into training set and validation set according to the ratio of 8:2, and the scattering curves to be deblurred are input into the deblurring UNet network model.

[0085] Instantiate the network model and input the number of data points of the point spread function. Select the Adam optimizer as the optimizer, set the learning rate to 0.0001, and train the samples in batches. The batch size of this training is set to 8, that is, 8 scattering curves are trained in one batch. The Epoch is set to 1000, and the MSE loss function is calculated for each Epoch. The loss value calculated by the MSE loss function is back-propagated to each node, and the training parameters are continuously updated through the Adam optimizer. During the model training process, after each Epoch is trained, the accuracy of the validation set is counted. If the current accuracy is greater than the historical optimal accuracy, the trained labels and UNet weight model are saved.

[0086] Figure 4 is one of the schematic diagrams of the small-angle scattering curve after deblurring provided in the embodiment of the present application, Figure 5 is one of the schematic diagrams of the point spread function prediction results provided in the embodiments of the present application. In one embodiment, the sample structure used is a hard nanosphere with a diameter of 50 nm. The scattering curve to be deblurred is input into the trained UNet weight model, and the deblurred scattering curve and the point spread function prediction result are output, respectively. Figure 4 and Figure 5 shown.

[0087] Figure 6 This is the second schematic diagram of the small-angle scattering curve after deblurring provided in the embodiment of the present application. Figure 72 is a schematic diagram of the point spread function prediction result provided in the embodiment of the present application. In another embodiment, the sample structure used is a layered stacking structure with 20 layers and a thickness of 100 nm. The scattering curve to be deblurred is input into the trained UNet weight model, and the deblurred scattering curve and the point spread function prediction result are output, respectively. Figure 6 and Figure 7 shown.

[0088] Figure 8 is a schematic diagram of the structure of a small-angle scattering measurement data deblurring device based on machine learning provided in an embodiment of the present application, such as Figure 8 As shown, the device at least includes:

[0089] An acquisition unit 801 is used to acquire a small-angle scattering curve to be deblurred;

[0090] The deblurring unit 802 is used to input the small-angle scattering curve to be deblurred into a pre-trained UNet weight model and output the deblurred small-angle scattering curve; the UNet weight model is constructed based on the deblurring UNet network model and the point spread function prediction UNet network model.

[0091] Fig. 9 is a schematic diagram of the structure of the deblurring unit provided in the embodiment of the present application, such as Fig. 9 As shown, the deblurring unit 802 specifically includes:

[0092] A deblurring UNet module 901 is used to generate a deblurring result based on the small-angle scattering curve to be deblurred;

[0093] A point spread function prediction UNet module 902 is used to generate a point spread function prediction result based on a one-dimensional array consisting of the number of point spread function data points corresponding to the small-angle scattering curve to be deblurred;

[0094] The convolution module 903 is used to perform a convolution operation based on the deblurring result and the point spread function prediction result to generate a deblurred small-angle scattering curve.

[0095] In some embodiments, the deblurring result is a real small-angle scattering curve after deblurring or a small-angle scattering curve after deblurring; the point spread function prediction result is a point spread function or a Gaussian parameter corresponding to the point spread function.

[0096] In some embodiments, the deblurring result is a deblurred small-angle scattering curve, and the convolution module 903 is specifically used for:

[0097] If the point spread function prediction result is a point spread function, the deblurred result is convolved with the point spread function prediction result after taking the natural exponent and taking the natural logarithm to generate a deblurred small-angle scattering curve; or,

[0098] If the point spread function prediction result is the Gaussian parameter corresponding to the point spread function, the point spread function is determined based on the Gaussian parameter, the deblurring result is convolved with the point spread function after taking the natural exponent and taking the natural logarithm to generate the deblurred small-angle scattering curve.

[0099] In some embodiments, the deblurring result is a real small-angle scattering curve after deblurring, and the convolution module 903 is specifically used for:

[0100] If the point spread function prediction result is a point spread function, a convolution operation is performed on the deblurring result and the point spread function prediction result and a natural logarithm is taken to generate a deblurred small-angle scattering curve; or,

[0101] If the point spread function prediction result is the Gaussian parameter corresponding to the point spread function, the point spread function is determined based on the Gaussian parameter, and a convolution operation is performed on the deblurring result and the point spread function and the natural logarithm is taken to generate a deblurred small-angle scattering curve.

[0102] In some embodiments, the deblurring UNet network model and the point spread function prediction UNet network model are both composed of multiple UNet networks connected in series.

[0103] In some embodiments, the activation function of the point spread function prediction UNet network model is a Softplus function.

[0104] In some embodiments, the weight model also outputs a point spread function prediction result.

[0105] In some embodiments, the UNet weight model is trained based on the following steps:

[0106] Obtaining small-angle scattering curves to be deblurred and after deblurring of different sample structures and different point spread functions as training data sets;

[0107] Based on the training data set, the deblurring UNet network model and the point spread function prediction UNet network model are trained respectively;

[0108] The trained UNet weight model is determined based on the trained deblurring UNet network model and the point spread function prediction UNet network model.

[0109] In some embodiments, the damage function of the UNet weight model is a mean square error loss function, and the optimizer adopts the Adam optimizer.

[0110] It is understandable that the detailed functional implementation of each of the above-mentioned units / modules can be found in the introduction of the above-mentioned method embodiment, and will not be described in detail here. It should be understood that the above-mentioned device is used to execute the method in the above-mentioned embodiment, and the corresponding program module in the device has a similar implementation principle and technical effect as described in the above-mentioned method. The working process of the device can refer to the corresponding process in the above-mentioned method, and will not be described in detail here.

[0111] Based on the method in the above embodiment, an embodiment of the present application provides an electronic device. The device may include: at least one memory for storing programs and at least one processor for executing the programs stored in the memory. When the program stored in the memory is executed, the processor is used to execute the method described in the above embodiment.

[0112] Fig.10 is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application, such as Fig.10 As shown, the electronic device may include: a processor 1001, a communication interface 1002, a memory 1003 and a communication bus 1004, wherein the processor 1001, the communication interface 1002 and the memory 1003 communicate with each other through the communication bus 1004. The processor 1001 may call the software instructions in the memory 1003 to execute the method described in the above embodiment.

[0113] In addition, the logic instructions in the above-mentioned memory 1003 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present application, or the part that contributes to the relevant technology or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application.

[0114] Based on the method in the above embodiment, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program runs on a processor, the processor executes the method in the above embodiment.

[0115] Based on the method in the above embodiment, an embodiment of the present application provides a computer program product. When the computer program product runs on a processor, the processor executes the method in the above embodiment.

[0116] It is understandable that the processor in the embodiment of the present application may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. The general-purpose processor may be a microprocessor or any conventional processor.

[0117] The method steps in the embodiments of the present application can be implemented by hardware, or by a processor executing software instructions. The software instructions can be composed of corresponding software modules, and the software modules can be stored in random access memory (Random Access Memory, RAM), flash memory, read-only memory (Read-Only Memory, ROM), programmable read-only memory (Programmable ROM, PROM), erasable programmable read-only memory (Erasable PROM, EPROM), electrically erasable programmable read-only memory (Electrically EPROM, EEPROM), registers, hard disks, mobile hard disks, CD-ROMs or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and can write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an ASIC.

[0118] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions can be transmitted from a website site, a computer, a server or a data center to another website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or a data center that includes one or more available media integration. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

[0119] It should be understood that the various numerical numbers involved in the embodiments of the present application are only used for the convenience of description and are not used to limit the scope of the embodiments of the present application.

[0120] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A method for deblurring a small-angle scattering curve based on machine learning, characterized in that: include: Obtaining a small-angle scattering curve to be deblurred; Inputting the small-angle scattering curve to be deblurred into a pre-trained UNet weight model, and outputting the deblurred small-angle scattering curve; the UNet weight model is constructed based on the deblurring UNet network model and the point spread function prediction UNet network model; The output of the deblurred small-angle scattering curve includes: Input the small-angle scattering curve to be deblurred into the deblurring UNet network model to generate a deblurring result; input a one-dimensional array consisting of the number of point spread function data points corresponding to the small-angle scattering curve to be deblurred into the point spread function prediction UNet network model to generate a point spread function prediction result; A convolution operation is performed based on the deblurring result and the point spread function prediction result to generate the deblurred small-angle scattering curve.

2. The small-angle scattering curve deblurring method according to claim 1, characterized in that: The deblurring result is a real small-angle scattering curve after deblurring or a small-angle scattering curve after deblurring; The point spread function prediction result is the point spread function or a Gaussian parameter corresponding to the point spread function.

3. The small-angle scattering curve deblurring method according to claim 2, characterized in that: The deblurring result is a small-angle scattering curve after deblurring, and generating the small-angle scattering curve after deblurring includes: If the point spread function prediction result is a point spread function, the deblurring result is convolved with the point spread function prediction result after taking the natural exponent and taking the natural logarithm to generate the deblurred small-angle scattering curve; or If the point spread function prediction result is a Gaussian parameter corresponding to the point spread function, the point spread function is determined based on the Gaussian parameter, the deblurring result is convolved with the point spread function after taking the natural exponent and taking the natural logarithm to generate the deblurred small-angle scattering curve.

4. The small-angle scattering curve deblurring method according to claim 2, characterized in that: The deblurring result is a real small-angle scattering curve after deblurring, and generating the small-angle scattering curve after deblurring includes: If the point spread function prediction result is a point spread function, performing a convolution operation on the deblurring result and the point spread function prediction result and taking a natural logarithm to generate the deblurred small-angle scattering curve; or, If the point spread function prediction result is a Gaussian parameter corresponding to the point spread function, the point spread function is determined based on the Gaussian parameter, a convolution operation is performed on the deblurring result and the point spread function, and a natural logarithm is taken to generate the deblurred small-angle scattering curve.

5. The small-angle scattering curve deblurring method according to claim 3, characterized in that: The activation function of the point spread function prediction UNet network model is the Softplus function.

6. The small-angle scattering curve deblurring method according to claim 1, characterized in that: The deblurring UNet network model and the point spread function prediction UNet network model are both composed of a plurality of UNet networks connected in series.

7. The small-angle scattering curve deblurring method based on machine learning according to claim 1, characterized in that: The weight model also outputs the point spread function prediction result.

8. The small-angle scattering curve deblurring method according to claim 1, characterized in that: The UNet weight model is trained based on the following steps: Obtaining small-angle scattering curves to be deblurred and after deblurring of different sample structures and different point spread functions as training data sets; Based on the training data set, respectively train the deblurring UNet network model and the point spread function prediction UNet network model; The trained UNet weight model is determined based on the trained deblurring UNet network model and the point spread function prediction UNet network model.

9. The small-angle scattering curve deblurring method according to claim 8, characterized in that: The damage function of the UNet weight model is the mean square error loss function, and the optimizer adopts the Adam optimizer.

10. A small-angle scattering curve deblurring device based on machine learning, characterized in that: include: An acquisition unit, used for acquiring a small-angle scattering curve to be deblurred; A deblurring unit, used for inputting the small-angle scattering curve to be deblurred into a pre-trained UNet weight model, and outputting the deblurred small-angle scattering curve; the UNet weight model is constructed based on the deblurring UNet network model and the point spread function prediction UNet network model; Wherein, the deblurring unit comprises: A deblurring UNet module, used for generating a deblurring result based on the small-angle scattering curve to be deblurred; A point spread function prediction UNet module is used to generate a point spread function prediction result based on a one-dimensional array consisting of the number of point spread function data points corresponding to the small-angle scattering curve to be deblurred; A convolution module is used to perform a convolution operation based on the deblurring result and the point spread function prediction result to generate the deblurred small-angle scattering curve.

Citation Information

Patent Citations

  • Method for removing blurring effect in X-ray scattering and diffraction experiments

    CN104268421A

  • Image deblurring method based on diffusion model

    CN116645287A

  • Image deblurring method and system based on improved UNet network

    CN116703786A

  • X-ray scattering device

    DE202021004378U1