Rare earth perovskite X-ray detector energy spectrum analysis method and system

Abstract

The invention relates to the technical field of energy spectrum analysis, and discloses a rare earth perovskite X-ray detector energy spectrum analysis method and system.The rare earth perovskite is irradiated through an X-ray detector, an original current signal sequence output by the detector under X-ray irradiation is collected, the original current signal sequence is preprocessed, and the energy spectrum of the X-ray detector is obtained; performing pulse shape discrimination on the preprocessed current signal, extracting a current pulse signal corresponding to a single X-ray photon event, extracting pulse characteristic parameters, inputting the pulse characteristic parameters into an energy spectrum analysis model, and mapping and outputting corresponding X-ray photon energy by the energy spectrum analysis model according to the input pulse characteristic parameters, according to the method, the unique response characteristic of the rare earth perovskite material to the X-ray energy is utilized, and the pulse shape discrimination technology and the machine learning model are combined, so that X-ray photons with different energies can be distinguished at high precision, and high-resolution energy spectrum analysis is realized.

Description

Technical Field

[0001] This invention relates to the field of energy spectrum analysis technology, and more specifically to energy spectrum analysis methods and systems for rare earth perovskite X-ray detectors. Background Technology

[0002] X-ray energy dispersive spectroscopy (EDS) plays a crucial role in fields such as medical imaging, security inspection, industrial non-destructive testing, and nuclear science. Traditional X-ray EDS systems typically employ semiconductor detectors such as silicon, germanium, or cadmium zinc telluride (CZT). However, these materials have some inherent drawbacks: silicon and germanium detectors require complex cryogenic cooling systems, are bulky, and inconvenient to use; while CZT materials exhibit excellent performance, their high manufacturing cost and difficult crystal growth limit their widespread application. In recent years, perovskite materials have shown great potential in the field of X-ray detection due to their advantages such as high atomic number, strong X-ray blocking ability, high carrier mobility-lifetime product, and ease of solution fabrication for low-cost, large-area devices. However, most current research on perovskite-based detectors focuses on improving detection sensitivity and reducing the detection limit. There is a lack of systematic methods and effective solutions for using perovskite detectors for high-precision energy spectrum analysis, especially for using their material properties to distinguish X-ray photons of different energies. Existing technologies struggle to achieve high-resolution, real-time energy spectrum analysis without significantly increasing system complexity and cost. Therefore, there is an urgent need for energy spectrum analysis methods and systems for rare-earth perovskite X-ray detectors. Summary of the Invention

[0003] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present invention provide a method and system for energy spectrum analysis of rare earth perovskite X-ray detectors to solve the technical problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a rare earth perovskite X-ray detector energy spectrum analysis method, comprising the following steps: Step S1: Irradiate rare earth perovskite with an X-ray detector and collect characteristic response current signals related to photon energy when irradiated with X-ray photons of different energies; Step S2: Acquire the raw current signal sequence output by the detector under X-ray irradiation; Step S3: Preprocess the original current signal sequence to obtain the preprocessed current signal; Step S4: Perform pulse shape discrimination on the preprocessed current signal to extract the current pulse signal corresponding to a single X-ray photon event; Step S5: Extract pulse characteristic parameters for each extracted current pulse signal; Step S6: Input the pulse characteristic parameters into the energy spectrum analysis model. The energy spectrum analysis model maps and outputs the corresponding X-ray photon energy according to the input pulse characteristic parameters. Step S7: Count the energy of all identified X-ray photons and generate an X-ray energy spectrum distribution.

[0005] In a preferred embodiment, a rare-earth perovskite X-ray detector energy spectrum analysis system includes an X-ray detector, an acquisition unit, a preprocessing unit, a discrimination unit, an extraction unit, an analysis unit, and an energy spectrum unit. The X-ray detector is used to irradiate rare-earth perovskite. The acquisition unit is used to acquire the raw current signal sequence output under X-ray irradiation. The preprocessing unit performs noise reduction and baseline correction on the raw current signal sequence. The discrimination unit processes the data after preprocessing. The extraction unit extracts current pulse signals. The analysis unit outputs X-ray photon energy. The energy spectrum unit generates an X-ray energy spectrum distribution map based on the X-ray photon energy.

[0006] In a preferred embodiment, the preprocessing unit includes a noise reduction module and a correction module. The noise reduction module performs noise reduction using wavelet transform. During wavelet transform noise reduction, the original signal is decomposed into different scales using wavelet basis functions to obtain a series of high-frequency and low-frequency wavelet coefficients. A threshold is set, and the wavelet coefficients are thresholded. Wavelet coefficients smaller than the threshold are set to zero, while coefficients larger than the threshold are retained. The processed wavelet coefficients are then subjected to inverse wavelet transform to reconstruct the noise-reduced current signal.

[0007] In a preferred embodiment, the correction module receives the reconstructed current signal and performs signal correction using dynamic baseline estimation. During dynamic baseline estimation, a time window slides across the signal, and the median or lower percentile of all data points within the window is calculated as the baseline estimate of the center point of the window. As the window slides, a dynamic baseline curve is generated. The noise-reduced current signal is subtracted from the baseline curve to complete the baseline correction.

[0008] In a preferred embodiment, when the discrimination unit extracts the current pulse signal, it sets a trigger time T, receives the preprocessed signal, compares the signal with an adjustable threshold, and records the point as a trigger point when the signal exceeds the threshold. The signal data within the time window of the trigger time T to be pushed forward and the trigger time T to be pushed backward is then extracted as the data to be measured.

[0009] In a preferred embodiment, the discrimination unit performs stacking analysis on the extracted test data. During the stacking analysis, the discrimination unit calculates the first derivative of the test data and records the number and position of the positive peak and the corresponding negative peak of the first derivative. When there are two or more positive peaks and corresponding negative peaks, the test data is a stacked pulse. When the pulse is stacked, the test data is discarded. When there is only one set of positive peaks and corresponding negative peaks, the test data is a current pulse signal.

[0010] In a preferred embodiment, the discrimination unit sends the current pulse signal to the extraction unit, the extraction unit extracts feature parameters including pulse amplitude, pulse rise time and pulse integral area from the current pulse signal, the extraction unit sends the extracted feature parameters to the analysis unit, the analysis unit combines the received feature parameters into a feature vector and inputs it into the energy spectrum analysis model, and the analysis unit outputs the X-ray photon energy through the energy spectrum analysis model.

[0011] In a preferred embodiment, the energy spectrum analysis model in the analysis unit is trained using a deep neural network model. During training, the energy spectrum analysis model is provided with a monoenergetic X-ray source of known energy. The rare-earth perovskite X-ray detector to be trained is placed under X-ray beam irradiation. Current pulse signals are collected for the selected monoenergetic X-ray source. For each pulse signal, its corresponding true energy value is labeled. All labeled data from different radiation sources are merged to form a total dataset. This dataset is randomly shuffled and divided into a training set, a validation set, and a test set in a 70:15:15 ratio.

[0012] In a preferred embodiment, each original pulse signal in the training set undergoes noise reduction and baseline correction, and feature parameters are extracted. These feature parameters form a feature vector. A deep neural network model is selected, and the feature vectors of all samples in the training set are used as input, while the corresponding labeled energy is used as the target output. The model is trained, and training is completed when the model loss function converges to a minimum value. The feature parameters in the training set are the same as the feature parameters extracted by the extraction unit.

[0013] In a preferred embodiment, the energy spectrum unit receives X-ray photon energy, groups and accumulates the received X-ray photon energy to generate energy spectrum histogram data, smooths the generated energy spectrum histogram data, and converts the horizontal axis from channel address to energy value according to the energy-channel address relationship, and plots and updates the energy spectrum distribution map in real time on the display device.

[0014] The technical effects and advantages of this invention are as follows: This invention utilizes the unique response characteristics of rare earth perovskite materials to X-ray energy, and combines pulse shape discrimination technology and machine learning models to accurately distinguish X-ray photons of different energies, achieving high-resolution energy spectrum analysis. This invention uses wavelet transform for noise reduction, which can effectively extract transient pulse signals from noise and preserve the sharp edges and shape features of the pulses. Baseline correction eliminates DC offset and low-frequency drift of the signal. After noise reduction and correction by the preprocessing unit, a current signal with high signal-to-noise ratio and stable baseline can be obtained, which is more accurate for subsequent pulse shape identification. This invention uses a discrimination unit to process current pulse signals, which can filter out environmental noise and dark current interference from the detector itself, ensuring that the signals analyzed subsequently are real X-ray events. Pulse accumulation is discarded to avoid distortion of energy spectrum analysis results, thus ensuring that this application can ultimately generate an accurate energy spectrum. This invention generates X-ray photon energy based on pulse amplitude, pulse rise time, and pulse integral area. Since the X-ray photon energy carries the characteristics of the material components that interact with it, the material can be detected when the final spectrum is formed, and the accuracy of the energy spectrum analysis is guaranteed. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the energy dispersive spectroscopy (EDS) analysis method of the present invention.

[0016] Figure 2 This is a schematic diagram of the energy dispersive spectroscopy (EDS) analysis system of the present invention. Detailed Implementation

[0017] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. In addition, the forms of the various structures described in the following embodiments are merely illustrative. The rare earth perovskite X-ray detector energy spectrum analysis method and system involved in the present invention are not limited to the structures described in the following embodiments. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Reference Figure 1 This invention provides a method for energy spectral analysis of rare-earth perovskite X-ray detectors, comprising the following steps: Step S1: Irradiate rare earth perovskite with an X-ray detector and collect characteristic response current signals related to photon energy when irradiated with X-ray photons of different energies; Step S2: Acquire the raw current signal sequence output by the detector under X-ray irradiation; Step S3: Preprocess the original current signal sequence to obtain the preprocessed current signal; Step S4: Perform pulse shape discrimination on the preprocessed current signal to extract the current pulse signal corresponding to a single X-ray photon event; Step S5: Extract pulse characteristic parameters for each extracted current pulse signal; Step S6: Input the pulse characteristic parameters into the energy spectrum analysis model. The energy spectrum analysis model maps and outputs the corresponding X-ray photon energy according to the input pulse characteristic parameters. Step S7: Calculate the energy of all identified X-ray photons and generate an X-ray energy spectrum distribution map.

[0019] Reference Figure 2 A rare-earth perovskite X-ray detector energy spectrum analysis system includes an X-ray detector, an acquisition unit, a preprocessing unit, a discrimination unit, an extraction unit, an analysis unit, and an energy spectrum unit. The X-ray detector is used to irradiate rare-earth perovskite. The acquisition unit is used to acquire the raw current signal sequence output under X-ray irradiation. The preprocessing unit performs noise reduction and baseline correction on the raw current signal sequence. The discrimination unit processes the data after preprocessing. The extraction unit extracts current pulse signals. The analysis unit outputs X-ray photon energy. The energy spectrum unit generates an X-ray energy spectrum distribution map based on the X-ray photon energy.

[0020] In this embodiment, by utilizing the unique response characteristics of rare earth perovskite materials to X-ray energy, i.e., photons of different energies generate current pulses of different shapes, and combining pulse shape discrimination technology and machine learning models, this application can distinguish X-ray photons of different energies with high precision, thus achieving high-resolution energy spectrum analysis.

[0021] Reference Figure 2 The preprocessing unit includes a noise reduction module and a correction module. The noise reduction module uses wavelet transform noise reduction. During wavelet transform noise reduction, the original signal is decomposed into different scales using wavelet basis functions to obtain a series of high-frequency and low-frequency wavelet coefficients. A threshold is set, and the wavelet coefficients are thresholded. Wavelet coefficients smaller than the threshold are set to zero, and coefficients larger than the threshold are retained. The processed wavelet coefficients are then subjected to inverse wavelet transform to reconstruct the noise-reduced current signal.

[0022] In this embodiment, wavelet transform denoising is used for noise reduction because it is one of the most advanced and effective methods for processing non-stationary signals. Wavelet transform denoising can simultaneously localize in both the time and frequency domains, effectively extracting transient pulse signals from noise while preserving the sharp edges and shape features of the pulses, facilitating subsequent pulse shape identification. Since the wavelet coefficients corresponding to the real signal are relatively large, while those corresponding to the noise are relatively small, a threshold is set when thresholding the wavelet coefficients. Coefficients smaller than the threshold are set to zero, while coefficients larger than the threshold are retained or reduced, effectively removing noise.

[0023] Reference Figure 2 The correction module receives the reconstructed current signal and performs signal correction using dynamic baseline estimation. During dynamic baseline estimation, a time window slides across the signal, and the median or lower percentile of all data points within the window is calculated as the baseline estimate of the center point of the window. As the window slides, a dynamic baseline curve is generated. The noise-reduced current signal is subtracted from the baseline curve to complete the baseline correction.

[0024] In this embodiment, baseline correction processing is used to eliminate DC offset and low-frequency drift of the signal. After noise reduction and correction processing by the preprocessing unit, a current signal with high signal-to-noise ratio and stable baseline can be obtained, which is more accurate when used for subsequent pulse shape discrimination.

[0025] Reference Figure 2 When the discrimination unit extracts the current pulse signal, it sets a trigger time T, receives the preprocessed signal, compares the signal with an adjustable threshold, and records the point when the signal exceeds the threshold as a trigger point. The signal data within the time window of the trigger time T forward and the trigger time T backward is extracted as the test data. The discrimination unit performs accumulation analysis on the extracted test data. During accumulation analysis, the discrimination unit calculates the first derivative of the test data and records the number and position of the positive peak and the corresponding negative peak of the first derivative. When there are two or more positive peaks and corresponding negative peaks, the test data is a piled pulse. The test data is discarded when the pulse is piled up. When there is only one set of positive peaks and corresponding negative peaks, the test data is a current pulse signal.

[0026] In this embodiment, when an X-ray beam irradiates the detector, it consists of a large number of photons with different energies and arrival times. The output of the detector is a continuous, fluctuating current signal, which is the result of countless photon events superimposed together. Its internal data is relatively chaotic. This application compares the signal with an adjustable threshold. When the signal exceeds the threshold, the point is recorded as the trigger point, and subsequent test data is generated. Signals below the threshold are regarded as noise. This can filter out environmental noise and dark current interference from the detector itself, ensuring that the signals analyzed later are real X-ray events. Pulse accumulation is discarded because accumulated pulses will distort and destroy the final generated X-ray energy spectrum, causing the energy spectrum analysis results to be completely distorted. This ensures that this application can accurately generate the energy spectrum in the end.

[0027] Reference Figure 2 The discrimination unit sends the current pulse signal to the extraction unit. The extraction unit extracts feature parameters including pulse amplitude, pulse rise time and pulse integral area from the current pulse signal. The extraction unit sends the extracted feature parameters to the analysis unit. The analysis unit combines the received feature parameters into a feature vector and inputs it into the energy spectrum analysis model. The analysis unit outputs the X-ray photon energy through the energy spectrum analysis model.

[0028] In this embodiment, X-ray photon energy based on pulse amplitude, pulse rise time, and pulse integral area can be generated. At this time, the X-ray photon energy carries the characteristics of the material components that interact with it. Therefore, when forming the final spectrum, the material can be detected and the accuracy of the energy spectrum analysis can be guaranteed.

[0029] Reference Figure 2 The energy spectrum analysis model in the analysis unit is trained using a deep neural network model. During training, a monoenergetic X-ray source with known energy is provided. The rare-earth perovskite X-ray detector to be trained is placed under X-ray beam irradiation. Current pulse signals are collected from the selected monoenergetic X-ray source. For each pulse signal, its corresponding true energy value is labeled. All labeled data from different radioactive sources are merged to form a total dataset. This dataset is randomly shuffled and divided into a training set, a validation set, and a test set in a 70:15:15 ratio. For each original pulse signal in the training set, noise reduction and baseline correction are performed, and feature parameters are extracted. These feature parameters form a feature vector. A deep neural network model is selected, and the feature vectors of all samples in the training set are used as input, with the corresponding labeled energy as the target output. Model training is performed. When the model loss function converges to a minimum value, training is complete. The feature parameters in the training set are the same as those extracted by the extraction unit.

[0030] In this embodiment, the energy spectrum analysis model obtained by means of this application is accurate enough to ensure the accuracy of rare earth perovskite identification. The energy spectrum analysis model adopts a multi-feature fusion strategy, and analyzes the amplitude, rise time, width, area and even the entire waveform of the pulse at the same time, which may more comprehensively reflect the true information of photon energy, thereby significantly improving energy resolution and accuracy.

[0031] Reference Figure 2 The energy spectrum unit receives X-ray photon energy, groups and accumulates the received X-ray photon energy to generate energy spectrum histogram data, smooths the generated energy spectrum histogram data, and converts the horizontal axis from channel address to energy value according to the energy-channel address relationship, and draws and updates the energy spectrum distribution map in real time on the display device.

[0032] In this embodiment, the energy spectrum unit converts a series of numbers representing photon energy into a clear energy spectrum distribution map with energy as the horizontal axis and count as the vertical axis. The energy spectrum map makes key information such as the characteristic peaks of elements and the absorption edges of materials readily apparent, allowing users to directly understand the data inside the rare earth perovskite X-ray detector. Furthermore, the energy spectrum distribution map is updated in real time to ensure detection accuracy.

[0033] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. The units and algorithm steps of the various examples described in the embodiments can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0034] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0035] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0036] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for energy spectral analysis of rare-earth perovskite X-ray detectors, characterized in that, Includes the following steps: Step S1: Irradiate rare earth perovskite with an X-ray detector and collect characteristic response current signals related to photon energy when irradiated with X-ray photons of different energies; Step S2: Acquire the raw current signal sequence output by the detector under X-ray irradiation; Step S3: Preprocess the original current signal sequence to obtain the preprocessed current signal; Step S4: Perform pulse shape discrimination on the preprocessed current signal to extract the current pulse signal corresponding to a single X-ray photon event; Step S5: Extract pulse characteristic parameters for each extracted current pulse signal; Step S6: Input the pulse characteristic parameters into the energy spectrum analysis model. The energy spectrum analysis model maps and outputs the corresponding X-ray photon energy according to the input pulse characteristic parameters. Step S7: Calculate the energy of all identified X-ray photons and generate an X-ray energy spectrum distribution map.

2. A rare-earth perovskite X-ray detector energy spectrum analysis system, characterized in that: The rare-earth perovskite X-ray detector energy spectrum analysis method as described in claim 1 includes an X-ray detector, an acquisition unit, a preprocessing unit, a discrimination unit, an extraction unit, an analysis unit, and an energy spectrum unit. The X-ray detector is used to irradiate rare-earth perovskite. The acquisition unit is used to acquire the raw current signal sequence output under X-ray irradiation. The preprocessing unit performs noise reduction and baseline correction on the raw current signal sequence. The discrimination unit processes the data after preprocessing. The extraction unit extracts current pulse signals. The analysis unit outputs X-ray photon energy. The energy spectrum unit generates an X-ray energy spectrum distribution map based on the X-ray photon energy.

3. The rare earth perovskite X-ray detector energy spectrum analysis system according to claim 2, characterized in that: The preprocessing unit includes a noise reduction module and a correction module. The noise reduction module uses wavelet transform noise reduction. During wavelet transform noise reduction, the original signal is decomposed into different scales using wavelet basis functions to obtain a series of high-frequency and low-frequency wavelet coefficients. A threshold is set, and the wavelet coefficients are thresholded. Wavelet coefficients smaller than the threshold are set to zero, and coefficients larger than the threshold are retained. The processed wavelet coefficients are then subjected to inverse wavelet transform to reconstruct the noise-reduced current signal.

4. The rare earth perovskite X-ray detector energy spectrum analysis system according to claim 3, characterized in that: The correction module receives the reconstructed current signal and performs signal correction using dynamic baseline estimation. During dynamic baseline estimation, a time window slides across the signal, and the median or lower percentile of all data points within the window is calculated as the baseline estimate of the center point of the window. As the window slides, a dynamic baseline curve is generated. The noise-reduced current signal is subtracted from the baseline curve to complete the baseline correction.

5. The rare-earth perovskite X-ray detector energy spectrum analysis system according to claim 2, characterized in that: When the discrimination unit extracts the current pulse signal, it sets a trigger time T, receives the preprocessed signal, compares the signal with an adjustable threshold, and records the point as the trigger point when the signal exceeds the threshold. The signal data within the time window of the trigger time T is pushed forward and the trigger time T is pushed backward is extracted as the data to be tested.

6. The rare-earth perovskite X-ray detector energy spectrum analysis system according to claim 5, characterized in that: The discrimination unit performs stacking analysis on the extracted test data. During the stacking analysis, the discrimination unit calculates the first derivative of the test data and records the number and position of the positive peak and the corresponding negative peak of the first derivative. When there are two or more positive peaks and corresponding negative peaks, the test data is a stacked pulse. When the pulse is stacked, the test data is discarded. When there is only one set of positive peaks and corresponding negative peaks, the test data is a current pulse signal.

7. The rare-earth perovskite X-ray detector energy spectrum analysis system according to claim 5, characterized in that: The discrimination unit sends the current pulse signal to the extraction unit. The extraction unit extracts feature parameters including pulse amplitude, pulse rise time and pulse integral area from the current pulse signal. The extraction unit sends the extracted feature parameters to the analysis unit. The analysis unit combines the received feature parameters into a feature vector and inputs it into the energy spectrum analysis model. The analysis unit outputs the X-ray photon energy through the energy spectrum analysis model.

8. The rare earth perovskite X-ray detector energy spectrum analysis system according to claim 7, characterized in that: The energy spectrum analysis model in the analysis unit is trained by a deep neural network model. During training, a monoenergetic X-ray source with known energy is provided. The rare earth perovskite X-ray detector to be trained is placed under X-ray beam irradiation. Current pulse signals are collected for the selected monoenergetic X-ray source. For each pulse signal, its corresponding true energy value is labeled. All labeled data from different radiation sources are merged to form a total dataset. The dataset is randomly shuffled and divided into a training set, a validation set, and a test set in a 70:15:15 ratio.

9. The rare-earth perovskite X-ray detector energy spectrum analysis system according to claim 8, characterized in that: For each original pulse signal in the training set, noise reduction and baseline correction are performed, and feature parameters are extracted. The feature parameters form a feature vector. A deep neural network model is selected, and the feature vectors of all samples in the training set are used as inputs, with the corresponding labeled energy as the target output. The model is trained, and training is completed when the model loss function converges to a minimum value. The feature parameters in the training set are the same as the feature parameters extracted by the extraction unit.

10. The rare earth perovskite X-ray detector energy spectrum analysis system according to claim 2, characterized in that: The energy spectrum unit receives X-ray photon energy, groups and accumulates the received X-ray photon energy to generate energy spectrum histogram data, smooths the generated energy spectrum histogram data, and converts the horizontal axis from channel address to energy value according to the energy-channel address relationship, and draws and updates the energy spectrum distribution map in real time on the display device.

Images