X-ray detection method and equipment based on perovskite material and medium

Abstract

The invention discloses an X-ray detection method and device based on a perovskite material and a medium, and relates to the technical field of signal processing, and the method comprises the steps: collecting a layered structure and a fluorescence signal, obtaining a structure parameter set, carrying out black field signal collection and multi-dose flat field calibration, and obtaining an initial non-uniformity correction parameter table; collecting a disturbance response signal, calculating a trap load indicating quantity, obtaining a trap state index and a suggested bias voltage adjusting quantity, and obtaining a stable working bias voltage and a trap state flag by adopting proportional-integral adjustment; collecting a photoelectric dual-channel signal, performing offset deduction and response correction according to the correction parameter table, and adaptively adjusting a fusion weight according to a trap state to obtain a fusion dose value and a pixel dose graph; and based on the fusion result, performing small-step iteration to update the correction parameter, calculating a detection quality index in combination with a trap state flag, and outputting a final dose graph. According to the invention, the fusion weight is adjusted according to the trap state through the photoelectric dual-channel signal, and the high-precision estimation of the fusion dose is realized.

Description

Technical Field

[0001] This invention relates to the field of signal processing technology, and in particular to an X-ray detection method, device and medium based on perovskite materials. Background Technology

[0002] With the widespread application of X-ray imaging and radiation dose monitoring technologies in medical imaging, non-destructive testing, and aerospace exploration, improving detector performance has gradually become a key research direction. Traditional silicon-based and cadmium telluride-based detectors have certain advantages in terms of high sensitivity, high linearity, and low noise response, but their fabrication costs are high and the processes are complex. In recent years, perovskite materials have become the core material direction for next-generation high-performance X-ray detectors due to their excellent photoelectric conversion efficiency, tunable bandgap characteristics, and low-temperature fabrication processes. Combined with intelligent signal processing algorithms, high-resolution imaging and dynamic dose self-calibration can be achieved.

[0003] However, existing perovskite X-ray detection methods still have two shortcomings. First, most schemes only use a single electrical channel signal for dose estimation, without considering carrier retention and signal drift caused by traps, resulting in insufficient response stability and large dose measurement errors. Second, existing methods mostly rely on static calibration parameters, making it difficult to achieve adaptive correction and online optimization during operation. Under long-term operation or complex irradiation conditions, non-uniformity errors accumulate, and the detection quality deteriorates, failing to meet the application requirements of high precision, stability, and self-calibration. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides an X-ray detection method based on perovskite materials to solve the problems of difficulty in correcting drift errors caused by traps due to single-channel signals in existing technologies, as well as the lack of an adaptive correction mechanism based on fusion residuals.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides an X-ray detection method based on perovskite materials, which includes: acquiring layered structure and fluorescence signals, obtaining a set of structural parameters, acquiring black field signals and multi-dose flat field calibration based on the set of structural parameters, and obtaining an initial non-uniformity correction parameter table. Collect disturbance response signals, calculate trap load indication, obtain trap status indicators and suggested bias voltage adjustment, and use proportional-integral adjustment of the working bias voltage based on the trap status indicators and suggested bias voltage adjustment to obtain stable working bias voltage and trap status indicators. The electrical channel signal and the optical channel signal are acquired. The electrical channel signal is biased and the response is corrected according to the initial non-uniformity correction parameter table. The fusion weight of the photoelectric channel is adjusted by the trap status flag to obtain the fusion dose value and pixel dose map. Based on the fused dose value and pixel dose map, the initial non-uniformity correction parameter table is updated in small steps to obtain the non-uniformity correction parameter table. By fusing dose values, non-uniformity correction parameter tables, and trap status indicators, a comprehensive set of detection quality indicators is calculated to obtain the final dose map.

[0007] As a preferred embodiment of the X-ray detection method based on perovskite materials described in this invention, the specific steps for acquiring the layered structure and fluorescence signal to obtain the structural parameter set are as follows: The energy spectrum of the X-ray source used was measured to obtain the target energy range, and the thickness of the absorption layer was selected within the target energy range. Based on the thickness of the absorption layer, the operating bias voltage is calculated, the ambient temperature is set, and the absorption layer thickness, operating bias voltage, and ambient temperature are combined to obtain the set of structural parameters.

[0008] As a preferred embodiment of the X-ray detection method based on perovskite materials described in this invention, the steps of acquiring black field signals and performing multi-dose flat-field calibration based on a set of structural parameters to obtain an initial non-uniformity correction parameter table are as follows: Under completely dark conditions, the X-ray source was turned off, and multiple frames of pixel output signals were collected, averaged, and pixel offset was obtained. The fluctuation range was statistically analyzed to obtain the pixel noise level. The pixel output signal is acquired, the pixel bias is subtracted, the response curve of the pixel output with dose change is obtained, the average output increase per unit dose is extracted as the gain coefficient, the nonlinear correction parameter is calculated for nonlinear pixels, and the pixel bias, pixel noise level, gain coefficient and nonlinear correction parameter are combined to obtain the initial non-uniformity correction parameter table.

[0009] In a preferred embodiment of the X-ray detection method based on perovskite materials described in this invention, the steps of acquiring the disturbance response signal, calculating the trap load indication, obtaining the trap status index and suggested bias adjustment are as follows: A small-amplitude sinusoidal disturbance is superimposed on the working bias voltage, the disturbance response signal is collected, the current fluctuation amplitude and phase delay are calculated frequency by frequency, and the frequency domain response spectrum is obtained. The trap load indication is calculated by the frequency domain response spectrum, the trap load reference baseline is set, the trap load indication is compared with the trap load reference baseline, the operating bias is adjusted, and when the trap inside the detector is in a stable equilibrium state, the trap status index is obtained and the suggested bias adjustment is generated.

[0010] In a preferred embodiment of the X-ray detection method based on perovskite materials described in this invention, the step of obtaining a stable operating bias and trap status indicators by adjusting the operating bias using proportional-integral methods based on trap state indicators and suggested bias adjustment amounts, specifically involves the following steps: Based on the trap condition indicators and suggested bias adjustment, the working bias is updated using proportional-integral adjustment. The current output signal is acquired in real time according to a fixed sampling period, the new trap load indication is calculated and the working bias is iteratively updated, and a fast reset is performed. The operating bias is restored to the stable point before reset, and the trap load indication is recalculated until the trap status flag is stable. The stable operating bias and trap status flag are then obtained.

[0011] As a preferred embodiment of the X-ray detection method based on perovskite materials described in this invention, the steps include: acquiring electrical channel signals and optical channel signals; performing bias subtraction and response correction on the electrical channel signals according to an initial non-uniformity correction parameter table; adjusting the photoelectric channel fusion weights through a trap state flag; and obtaining the fused dose value and pixel dose map. The photocurrent signal of the electrical channel is collected to form an electrical signal matrix, and the fluorescence signal of the light-transmitting window is collected to form an optical signal matrix. The electrical signal matrix is ​​subjected to bias subtraction, gain correction and nonlinear compensation based on the initial non-uniformity correction parameter table to obtain the electrical channel correction signal; The electrical channel correction signal and the optical signal matrix are normalized to obtain pixel-level electrical channel dose estimates and optical channel dose estimates. The photoelectric fusion weights are calculated and weighted to obtain the fused dose value. The fused dose values ​​of all pixels are combined according to their spatial location to form a two-dimensional pixel dose map, thus obtaining the pixel dose map.

[0012] As a preferred embodiment of the X-ray detection method based on perovskite materials described in this invention, the step of iteratively updating the initial non-uniformity correction parameter table based on the fused dose value and pixel dose map to obtain the non-uniformity correction parameter table includes the following specific steps: Based on the fused dose value and pixel dose map, the target dose reference value is determined in a flat field or known dose distribution scene, and the residual of each pixel is calculated. When the trap status flag is stable, the pixel bias, gain coefficient and nonlinear correction parameter are iteratively updated in small steps using the residual of each pixel to obtain the parameter update amount. The update is paused when the trap status flag is that the trap is overfilled or underfilled. Based on the parameter update amount, spatial smoothing constraints are applied to the update amounts of adjacent pixels, and periodic cache accumulation is performed to obtain the non-uniformity correction parameter table.

[0013] As a preferred embodiment of the X-ray detection method based on perovskite materials described in this invention, the specific steps for calculating a comprehensive detection quality index set and obtaining the final dose map by fusing dose values, non-uniformity correction parameter tables, and trap status flags are as follows: Based on the fusion dose value, non-uniformity correction parameter table and trap status flag, flat field, oblique target and black field and low dose flat field data are collected sequentially to obtain flat field fusion dose map, oblique target fusion dose map and black field and low dose flat field sequence. Spatial non-uniformity index is calculated by flat-field fusion dose map, edge spread function and line spread function are extracted by oblique-side target fusion dose map to obtain MTF index, and noise equivalent dose index is obtained by statistically analyzing noise in black field and low-dose flat field sequences. Quality compliance is determined based on spatial non-uniformity index, MTF index, noise equivalent dose index, and trap status indicator. When the spatial non-uniformity index, MTF index, and noise equivalent dose index all meet the standards and the trap status indicator is stable, the final fusion dose map is obtained.

[0014] In a second aspect, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, it implements any step of the X-ray detection method based on perovskite materials as described in the first aspect of the present invention.

[0015] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein, when the computer program is executed by a processor, it implements any step of the X-ray detection method based on perovskite materials as described in the first aspect of the present invention.

[0016] The beneficial effects of this invention are as follows: through a highly stable layered structure, high-sensitivity detection and low-noise output are achieved within the target X-ray energy range; by establishing an initial non-uniformity correction parameter table combined with trap sensing and proportional-integral bias adjustment strategies, dynamic correction of pixel-level response consistency and trap stability is achieved; and by adaptively adjusting the fusion weights based on the trap state of the photoelectric dual-channel signal, high-precision estimation of the fusion dose and online optimization of detector performance are achieved. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1This is a flowchart of an X-ray detection method based on perovskite materials.

[0019] Figure 2 This is a flowchart for the calibration and trap state control of perovskite detectors.

[0020] Figure 3 This is a flowchart of dual-channel signal acquisition and fusion dose estimation.

[0021] Figure 4 A flowchart for integrating residual-driven parameter updates and detection quality assessment. Detailed Implementation

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0025] Reference Figures 1-4 This is one embodiment of the present invention, which provides an X-ray detection method based on perovskite materials, comprising the following steps: S1: Acquire layered structure and fluorescence signals, obtain structural parameter set, and based on the structural parameter set, acquire black field signal and perform multi-dose flat field calibration to obtain initial non-uniformity correction parameter table.

[0026] Based on the target X-ray energy range, select a thickness that ensures the absorption layer absorbs at least 90% of the target energy X-rays.

[0027] It should be noted that the target X-ray energy range is determined by measuring the energy spectrum of the X-ray source used, combining the main energy concentration ranges in typical medical imaging and industrial testing, and selecting the range with a cumulative energy contribution of more than 90% as the design target energy range. The thickness is selected based on a high absorption rate to ensure signal strength, the ability of charge carriers to completely drift to the electrode under bias driving, and the film thickness that can be achieved by the process.

[0028] The target electric field strength is set to 2×104 V / cm ensures that electrons and holes can drift to the electrode within hundreds of nanoseconds when the absorption layer thickness is 150μm, avoiding signal loss caused by recombination. The working bias voltage of 30V can be deduced from the target electric field strength and the absorption layer thickness.

[0029] To reduce the impact of ambient temperature fluctuations on carrier mobility and trap excitation, the device temperature is kept stable at 25°C.

[0030] It should be noted that the device temperature is kept stable at 25℃ because 25℃ is the optimal performance point for perovskite materials at room temperature. This ensures that charge carriers can completely drift to the electrodes during the lifetime and effectively suppresses signal drift caused by dark current rise and trap activation.

[0031] The absorption layer thickness, target electric field strength, operating bias voltage, and device temperature are combined to obtain a set of structural parameters.

[0032] Under completely dark conditions, the X-ray source is turned off, the detector is in a radiation-free state, the output signals of all pixels are collected, and several frames (e.g., 100 frames) are recorded continuously. The multi-frame output of each pixel is averaged to obtain the pixel offset. The fluctuation range of the multi-frame output of the pixel is statistically analyzed to obtain the pixel noise level.

[0033] Turn on the X-ray source and adjust the output power. Under the premise of ensuring uniform irradiation field distribution, apply multiple known dose levels (e.g., 0.1 mGy / s, 0.5 mGy / s, 1.0 mGy / s, 2.0 mGy / s, 3.0 mGy / s) step by step. Maintain a constant irradiation time at each dose level, collect the output signal of each pixel, subtract the pixel bias at each dose level, and then compare the pixel output at different dose levels to obtain the response curve of pixel output as a function of dose.

[0034] For pixels whose response changes approximately linearly with the dose, the average output increase per unit dose is used as the gain coefficient. For pixels whose response deviates slightly from linearity in the high-dose range, the nonlinear response trend is extracted by comparing the growth rates in different dose ranges, and the nonlinear correction parameters are obtained.

[0035] The pixel offset, pixel noise level, gain coefficient, and nonlinear correction parameters of each pixel are combined to obtain the initial non-uniformity correction parameter table.

[0036] S2: Collect disturbance response signals, calculate trap load indication, obtain trap status indicators and suggested bias voltage adjustment, and adjust the operating bias voltage using proportional-integral method based on the trap status indicators and suggested bias voltage adjustment to obtain stable operating bias voltage and trap status indicators.

[0037] To analyze the trap response characteristics, a small-amplitude sinusoidal perturbation signal with an amplitude of ±1V is superimposed on the reference bias voltage. Multiple frequency points (e.g., 10Hz, 100Hz, 1kHz, 10kHz) are selected as perturbation frequencies, applied sequentially, each lasting 1 second with a 0.5-second interval. The perturbation signal is superimposed onto the bias voltage terminal through a precision signal source, forming a time-varying bias voltage input. Within each frequency perturbation cycle, the corresponding perturbation response signal is acquired, and analyzed to calculate the current fluctuation amplitude, extract the phase delay, and obtain the frequency domain response spectrum. Specifically, the complete current-time waveform of the detector at the current frequency is recorded. The sampled current data is statistically analyzed over one or more complete cycles, and the difference between the maximum and minimum values ​​of the current waveform is calculated. Half of this difference is taken to obtain the current fluctuation amplitude corresponding to the frequency. The voltage and current signals are subjected to Fast Fourier Transform (FFT) respectively to obtain their complex spectra in the frequency domain. The main peak components of the corresponding frequencies are extracted from the complex spectra, and the phase angles of the voltage and current signals are recorded. Then, using the phase angle of the voltage signal as a reference, the offset of the phase angle of the current signal is calculated, which is the phase delay at the current frequency. The current fluctuation amplitude and phase delay data obtained at various disturbance frequencies are organized and plotted with the disturbance frequency as the abscissa and the current response amplitude as the ordinate to construct a curve of current response amplitude versus frequency. The amplitude data at different frequency points are normalized, and a smoothing filtering algorithm is used to form a continuous response curve, thus obtaining the frequency domain response spectrum.

[0038] To quantitatively characterize the degree of trap filling, a trap load indicator is defined, expressed as: ; in, Indicates the trap load indication value. This indicates the upper limit of the perturbation frequency. This indicates the lower limit of the perturbation frequency. Indicates the detector at frequency The corresponding current response amplitude, Indicates frequency The corresponding reference current response amplitude.

[0039] It should be noted that, This is the reference response amplitude obtained by the detector under steady-state conditions where the trap reaches equilibrium, through multi-frequency perturbation bias voltage testing. The upper and lower limits of the perturbation frequency are determined by initially scanning the detector's current response at different frequencies to observe the changing trends of current amplitude and phase delay. When the frequency is too low (e.g., below 1 Hz), the trap's charging and discharging process completely follows the bias voltage changes, and the response difference is not significant. When the frequency is too high (e.g., above 10 kHz), the trap cannot respond promptly, and the current change tends to stabilize. Therefore, an effective frequency band covering the dynamic process of trap charging and discharging is selected as the perturbation frequency range. The minimum frequency at which the corresponding trap can respond adequately. The critical frequency at which the corresponding trap cannot fully respond ensures that the integration interval can comprehensively characterize the dynamic characteristics of the trap.

[0040] Compare the trap load indication with the trap load reference baseline. When the trap load indication is basically consistent with the trap load reference baseline, it indicates that the traps inside the detector are in a stable equilibrium state, the carrier charging and discharging process is normal, and no bias adjustment is required. If the trap load indication is significantly higher than the trap load reference baseline, it indicates that the traps are overfilled, with a large number of carriers remaining, which may cause output signal drift or response delay. In this case, the bias voltage should be appropriately increased to enhance the carrier drift capability and accelerate trap release. If the trap load indication is significantly lower than the trap load reference baseline, it indicates that the traps are not fully filled, and the device is in an undersaturated state, which may cause transient response instability or baseline drift. In this case, the bias voltage can be reduced or the steady-state operating time can be extended to promote gradual trap filling, achieve stable equilibrium, and obtain trap status indicators.

[0041] If the traps are overfilled, it is recommended to increase the operating bias voltage by 1-2V. If the traps are underfilled, it is recommended to decrease the operating bias voltage by 1-2V. If the voltage is stable, maintain the existing operating bias voltage to obtain a stable operating bias voltage.

[0042] It should be noted that adjusting the operating bias voltage to 1-2V is determined based on the response characteristics between the device's electric field strength and the trap release rate. Under isothermal conditions, the operating bias voltage is gradually increased, and the trend of the trap load indication is monitored. It was found that when the bias voltage adjustment range is within 3% to 7% of the reference value (corresponding to about 1-2V), the trap release rate can be significantly improved, and the load indication recovers to the steady-state level within a few minutes without causing excessive electric field stress or an increase in device leakage current. If the adjustment range is too small (less than 1V), the electric field change is insufficient to drive the trap release, and the trap retention problem cannot be effectively alleviated. If the adjustment range is too large (more than 2V), it may lead to local overheating, increased current noise, or even electrochemical instability of the perovskite layer.

[0043] A proportional-integral control algorithm is used to adjust the operating bias voltage in real time according to the trap load deviation. The expression is: ; in, This indicates the updated operating bias. Indicates the current working bias. Represents the proportional gain coefficient. Represents the integral gain coefficient. Indicates the trap load deviation. Indicates the current moment. Indicates the start time of the current control cycle. It should be noted that, This represents the difference between the trap load indication and the trap load reference baseline. The trap load reference baseline is determined by monitoring the trend of the trap load indication over a long period of operation. When the results of multiple consecutive tests stabilize within a certain fixed range, the average value is automatically recorded as the trap load reference baseline. The optimal value is obtained by selecting the value that can bring the trap load deviation to converge within several control cycles under the condition of no oscillation. It is set according to the recovery time characteristics of the device trap state, and is used to eliminate long-term deviations and make the trap state balanced and stable. The value range is 0.05 to 0.2V / second. A value less than 0.05 will cause residual deviation, and a value greater than 0.2 will cause bias voltage fluctuation.

[0044] The current output signal is acquired in real time with a fixed sampling period (e.g., 1 second), the new trap load indication is calculated, and the operating bias is updated in each cycle. To prevent the continuous accumulation of trap carriers during long-term operation, which would lead to response drift and decreased stability, the change in trap load deviation is continuously monitored during bias control. When the absolute value of the current trap load deviation exceeds 20% of the trap load reference baseline, or when the deviation fails to converge to the stable threshold for several consecutive control cycles (e.g., 5 cycles), it is determined that the trap state has significantly deviated from equilibrium and cannot be restored by conventional adjustment in a short time. A rapid reset is then performed by applying a short-duration high-voltage pulse to the detector to provide an instantaneous enhanced electric field. The high electric field driving effect causes the carriers trapped in the trap to be released rapidly. After the pulse ends, the operating bias is restored to the previous stable operating point, and the trap load indication is monitored again until the indication returns to the stable range. The trap state flag is then updated to stable, and the updated stable operating bias and trap state flag are obtained.

[0045] It should be noted that when the absolute value of the trap load deviation is less than 20% of the trap load reference baseline, it can be smoothly eliminated through conventional bias adjustment. When the deviation exceeds 20% of the trap load reference baseline, the conventional closed-loop adjustment response is slow and difficult to recover in a short time. A rapid reset needs to be triggered to achieve instantaneous correction. The stability threshold is the allowable residual range of the trap load deviation, reflecting the criteria for determining whether the trap state tends towards equilibrium. It is determined by analyzing the natural fluctuation characteristics under equilibrium conditions and is usually set to ±5% of the trap load reference baseline. When the deviation is less than the stability threshold, the trap state is considered stable.

[0046] S3: Acquire electrical channel signals and optical channel signals, perform bias subtraction and response correction on electrical channel signals according to the initial non-uniformity correction parameter table, adjust the photoelectric channel fusion weight through the trap status flag, and obtain the fusion dose value and pixel dose map.

[0047] The photocurrent signal generated by the perovskite absorption layer is acquired through the electrical channel to form an electrical signal matrix, and the fluorescence signal output from the light-transmitting window is acquired through the optical channel to form an optical signal matrix. Each pixel position... For the corresponding electrical and optical signal channels, pixel-level processing is performed on the electrical channel signal based on the initial non-uniformity correction parameter table, as expressed by: ; in, Indicates the electrical channel correction signal. This represents the original electrical channel output signal value. Indicates the pixel bias signal value. Represents the pixel gain coefficient. This represents the nonlinear correction coefficient.

[0048] It should be noted that, Under stable operating bias and constant temperature conditions, ensuring the detector is free from X-ray irradiation (i.e., the optical path is completely blocked and ambient light interference is shielded), several frames of electrical channel signals are continuously acquired, the output signal value at each pixel position is recorded, and the pixel bias signal value is obtained by averaging multiple frames of signals from the same pixel. This method employs a multi-dose flat-field irradiation approach, applying several known dose levels (e.g., 0.5 mGy, 1.0 mGy, 2.0 mGy) to the detector. The electrical channel output signal of each pixel is acquired, and subtracted using the pixel bias signal value to obtain the effective response value at each dose. A linear relationship is established between the effective response value and the corresponding dose, and the slope is fitted using the least squares method to obtain the pixel's response sensitivity, i.e., the pixel gain coefficient. The pixel gain coefficient ranges from [0.85, 1.15]. Values ​​exceeding this range typically indicate abnormal pixel response or process defects. This method avoids over-amplification to correct noise and facilitates the shielding of defective pixels. The detector is subjected to flat-field irradiation with multiple doses of different intensity. The electrical channel output signals of the pixels at each intensity are collected, and the pixel bias signal value is subtracted to obtain the effective response value of the pixel under different doses. The incident dose is obtained by dose calibration of the X-ray source. The effective response value is compared with the incident dose to observe the deviation of the response curve from the ideal linear relationship. Based on the deviation magnitude, the response bending trend of the pixel in the high-dose region is determined, and a correction coefficient that can compensate for nonlinear deviation is calculated. The nonlinear correction coefficient is obtained, and its value range is [0, 0.08]. A value less than 0 will cause reverse compensation and amplify noise, while when When the value exceeds 0.08, the magnitude of the quadratic term correction is too large, which may introduce overcompensation and signal distortion, affecting dose linearity and imaging stability.

[0049] Under stable operating bias and constant temperature conditions, flat-field irradiation with several known standard doses (e.g., 0.2, 0.5, 1.0, 2.0 mGy) is applied, and the electrical channel correction signal and optical channel normalized signal are collected respectively. The correspondence between the output signal of each channel and the actual incident dose is established. With the known dose as the horizontal axis and the channel signal as the vertical axis, each sampling point is fitted to form the dose response curves of the electrical channel and optical channel respectively, reflecting the law of change of each channel signal with the incident dose. During the operation phase, the real-time collected corrected electrical signal and normalized optical signal are substituted into their respective dose response curves, and the pixel-level electrical channel dose estimate and optical channel dose estimate are obtained by looking up the table, realizing the quantitative dose mapping of the photoelectric dual channels.

[0050] Based on the trap status flag and trap load deviation, the photoelectric channel weight is dynamically adjusted, as expressed by: ; ; in, Indicates the electrical channel weight. Indicates the optical channel weights. This indicates the adjustment sensitivity coefficient.

[0051] It should be noted that, It combines the trap load deviation The statistical relationship between the electrical channel signal drift and the signal drift is analyzed when... The contribution error of the electrical signal to the fusion dose value is increased. Then, based on the sensitivity of the drift error, the required suppression ratio is determined so that when the trap deviation is large, the electrical channel weight is significantly reduced and the optical channel weight is correspondingly increased, thereby maintaining the stability of the fusion dose. By establishing a linear response relationship between the rate of change of electrical channel weight and the trap deviation, the ratio that minimizes the variance of the fusion result is selected. The value is used as the adjustment sensitivity coefficient.

[0052] The expression for obtaining the fused dose value and fusion confidence is: ; ; in, Indicates the fusion dose value, This represents the pixel-level electrical channel dose estimate. This represents the pixel-level optical channel dose estimate. Indicates the fusion confidence level. This represents the stability coefficient.

[0053] It should be noted that, This study analyzes the inter-frame fluctuation amplitude of the fusion dose output under different trap load deviation levels by fitting the data. It calculates the response relationship between the fusion dose standard deviation and the deviation level, and selects the coefficient that minimizes the fitting error between the fusion confidence change and the output uncertainty as the optimal value. .

[0054] A pixel dose map is obtained by constructing a two-dimensional spatial distribution map using the fused dose values ​​of all pixel locations.

[0055] S4: Based on the fused dose value and pixel dose map, the initial non-uniformity correction parameter table is updated in small steps to obtain the non-uniformity correction parameter table.

[0056] The residual between the fused dose value and the target dose reference value is calculated to obtain the pixel residual reflecting the correction error. A slow-converging, small-step iterative approach is used to update the pixel bias, gain coefficient, and nonlinear correction coefficient in the initial non-uniformity correction parameter table. The expression is as follows: ; in, Indicates pixel residual. Indicates the first Pixel at the next iteration A certain correction parameter, including pixel offset, gain coefficient, and nonlinearity correction coefficient, Indicates the first Pixel at the next iteration A certain correction parameter, This indicates that the step size coefficient is being updated.

[0057] It should be noted that, Under stable operating conditions, small perturbations (e.g., 1% change) are applied to the pixel bias, gain coefficient, and nonlinear correction coefficient, respectively. The changes in the corresponding pixel fusion dose residuals are observed. The ratio of the residual change to the parameter change is used as the parameter sensitivity index to the output. The sensitivity is determined based on the convergence speed of the pixel residuals and the parameter adjustment sensitivity. The target dose reference value is a reference scene where the dose distribution can be determined under known irradiation conditions (e.g., flat field irradiation). In flat field irradiation, the global average fusion dose can be directly used as the reference value to make the dose of each pixel equal under ideal conditions.

[0058] When the trap status flag is stable, parameter updates are performed. When the trap status flag is that the trap is overfilled or underfilled, updates are paused to prevent trap drift from causing incorrect corrections. The parameter update amount of adjacent pixels is subject to smooth constraints to avoid local over-adjustment leading to spatial noise. The update results of each cycle are temporarily cached, and after accumulating several cycles, a non-uniformity correction parameter table is generated synchronously.

[0059] S5: By fusing dose values, non-uniformity correction parameter tables, and trap status flags, calculate the comprehensive detection quality index set and obtain the final dose map.

[0060] Under conditions of no obstruction, the detection surface is uniformly irradiated using a standard X-ray source to achieve a uniform dose (e.g., 1 mGy). By controlling the acquisition sequence, single or multiple frames of fused dose output images are recorded and averaged to obtain a flat-field fused dose map. A standard metal beveled target or resolution strip target is placed in front of the detection surface, and the angle between the target and the pixel matrix is ​​adjusted. Several frames of images are acquired under irradiation conditions consistent with the flat field to obtain the beveled target fused dose map. The X-ray source is turned off, and only the working bias is maintained. Multiple frames of black field output are acquired, and the dark noise level is statistically analyzed. The irradiation dose is reduced to 5%–10% of the uniform dose, and multiple frames of images are continuously acquired. The slope of the linear region between the output and the dose response is analyzed to obtain the black field and low-dose flat-field sequences.

[0061] It should be noted that the acquisition timing is controlled by pre-setting acquisition parameters such as exposure time, number of frames, interval and trigger delay, so that X-ray exposure and detector readout are synchronized under a unified clock, ensuring that each frame of image corresponds to the same irradiation conditions and integration time.

[0062] Using the fusion dose values ​​of all pixels in the flat-field fusion dose map as samples, calculate the maximum, minimum, and average values ​​of the fusion dose values. The difference between the maximum and minimum values ​​is taken as the fusion dose value difference, and the ratio of the fusion dose value difference to the average value is taken as the spatial non-uniformity index.

[0063] In the fusion dose map of the oblique target, the region containing the oblique edge is selected, and the gray-level distribution along the direction perpendicular to the edge is extracted. The edge spread function (ESF) is formed by subpixel resampling, and the derivative of the ESF is used to obtain the line spread function (LSF). The LSF is then subjected to Fourier transform to obtain the modulation transfer function (MTF) curve in the frequency domain.

[0064] In the black field and low-dose flat field sequences, the noise variance of the fused dose map is statistically analyzed, and the noise equivalent dose is calculated by the low-dose response slope (the rate of change of the fused dose value relative to the dose). This is the dose size corresponding to a signal-to-noise ratio of 1, which is used to measure the system's lowest detectable dose capability.

[0065] The spatial non-uniformity index, MTF, and noise equivalent dose are combined to obtain a comprehensive three-indicator. To integrate the three-indicator and trap load deviation on a single scalar, compliance criteria are set for each. When the spatial non-uniformity index is less than or equal to the uniformity tolerance threshold, the measured value of MTF at the reference frequency is greater than or equal to the lower limit of contrast requirement, and the noise equivalent dose is less than or equal to the compliance threshold of noise equivalent dose, it indicates that the comprehensive three-indicator meets the requirements. Simultaneously, the trap load deviation is read and compared with 20% of the trap load reference baseline. If the deviation is within the allowable range, the trap state is considered stable. Only when all three indicators meet the standard and the trap state is stable is the measurement quality considered qualified, and the final fused dose map is output. If the spatial non-uniformity indicator does not meet the standard, residual-driven small-step iterative updates of pixel bias, gain coefficient, and nonlinear correction parameters are enabled and take effect when the trap state indicator is stable. If the MTF does not meet the standard, the working bias is slightly increased to enhance the drift field and compress lateral diffusion. If the noise equivalent dose does not meet the standard, these pixel biases are adjusted to their multi-frame average value to eliminate fixed offset error. For pixels with sensitivity deviation, the gain coefficient is finely adjusted by statistically analyzing the ratio of fused dose to uniform dose to make the pixel response more consistent. If the trap state indicator indicates that the trap is overfilled or underfilled, a multi-frequency perturbation bias test is performed and a fast reset is executed until the trap state indicator is stable.

[0066] It should be noted that the uniformity tolerance threshold is determined by statistically analyzing the spatial non-uniformity of multiple standard detector outputs under flat-field illumination conditions, and taking the upper limit of its 95% confidence interval as the uniformity tolerance threshold. The lower limit of contrast requirement is determined by measuring the MTF response curves at a reference frequency under different contrast ratios according to the expected spatial resolution requirements, and taking the minimum MTF value that meets the condition of clear and discernible image quality as the lower limit. The threshold for meeting the noise equivalent dose is obtained by setting the minimum detectable dose requirement under the condition of minimum acceptable signal-to-noise ratio (e.g., SNR=1), and then inversely calculating it by combining the noise level and the response slope.

[0067] This embodiment also provides a computer device applicable to X-ray detection methods based on perovskite materials, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the X-ray detection method based on perovskite materials as proposed in the above embodiment.

[0068] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0069] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the X-ray detection method based on perovskite materials as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0070] In summary, this invention achieves high-sensitivity detection and low-noise output within the target X-ray energy range by selecting a lead-free perovskite absorption layer and setting thickness, electric field strength, and isothermal conditions that satisfy high absorptivity and high drift efficiency, thus forming a highly stable layered structure. Furthermore, by establishing an initial non-uniformity correction parameter table for pixel bias, gain, and nonlinear response under black field and multi-dose flat field conditions, and combining trap sensing and proportional-integral bias adjustment strategies, dynamic correction of pixel-level response consistency and trap stability is achieved. Finally, by acquiring photoelectric dual-channel signals and adaptively adjusting the fusion weight according to the trap state, combined with small-step iterative parameter updates driven by the fusion dose residual, high-precision estimation of the fusion dose and online self-optimization of detector performance are achieved.

[0071] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. An X-ray detection method based on perovskite materials, characterized in that: include, The layered structure and fluorescence signal were collected to obtain a set of structural parameters. Based on the set of structural parameters, black field signal acquisition and multi-dose flat field calibration were performed to obtain an initial non-uniformity correction parameter table. Collect disturbance response signals, calculate trap load indication, obtain trap status indicators and suggested bias voltage adjustment, and use proportional-integral adjustment of the working bias voltage based on the trap status indicators and suggested bias voltage adjustment to obtain stable working bias voltage and trap status indicators. The electrical channel signal and the optical channel signal are acquired. The electrical channel signal is biased and the response is corrected according to the initial non-uniformity correction parameter table. The fusion weight of the photoelectric channel is adjusted by the trap status flag to obtain the fusion dose value and pixel dose map. Based on the fused dose value and pixel dose map, the initial non-uniformity correction parameter table is updated in small steps to obtain the non-uniformity correction parameter table. By fusing dose values, non-uniformity correction parameter tables, and trap status indicators, a comprehensive set of detection quality indicators is calculated to obtain the final dose map.

2. The X-ray detection method based on perovskite materials as described in claim 1, characterized in that: The specific steps for acquiring the layered structure and fluorescence signal to obtain the structural parameter set are as follows: The energy spectrum of the X-ray source used was measured to obtain the target energy range, and the thickness of the absorption layer was selected within the target energy range. Based on the thickness of the absorption layer, the operating bias voltage is calculated, the ambient temperature is set, and the absorption layer thickness, operating bias voltage, and ambient temperature are combined to obtain the set of structural parameters.

3. The X-ray detection method based on perovskite materials as described in claim 2, characterized in that: The process of acquiring black field signals and performing multi-dose flat-field calibration based on the structural parameter set to obtain an initial non-uniformity correction parameter table involves the following steps: Under completely dark conditions, the X-ray source was turned off, and multiple frames of pixel output signals were collected, averaged, and pixel offset was obtained. The fluctuation range was statistically analyzed to obtain the pixel noise level. The pixel output signal is acquired, the pixel bias is subtracted, the response curve of the pixel output with dose change is obtained, the average output increase per unit dose is extracted as the gain coefficient, the nonlinear correction parameter is calculated for nonlinear pixels, and the pixel bias, pixel noise level, gain coefficient and nonlinear correction parameter are combined to obtain the initial non-uniformity correction parameter table.

4. The X-ray detection method based on perovskite materials as described in claim 3, characterized in that: The specific steps for acquiring the disturbance response signal, calculating the trap load indication, obtaining the trap status index and suggested bias voltage adjustment are as follows: A small-amplitude sinusoidal disturbance is superimposed on the working bias voltage, the disturbance response signal is collected, the current fluctuation amplitude and phase delay are calculated frequency by frequency, and the frequency domain response spectrum is obtained. The trap load indication is calculated by the frequency domain response spectrum, the trap load reference baseline is set, the trap load indication is compared with the trap load reference baseline, the operating bias is adjusted, and when the trap inside the detector is in a stable equilibrium state, the trap status index is obtained and the suggested bias adjustment is generated.

5. The X-ray detection method based on perovskite materials as described in claim 4, characterized in that: The process involves adjusting the operating bias using proportional-integral methods based on the trap status index and suggested bias adjustment amount to obtain a stable operating bias and trap status flag. The specific steps are as follows: Based on the trap condition indicators and suggested bias adjustment, the working bias is updated using proportional-integral adjustment. The current output signal is acquired in real time according to a fixed sampling period, the new trap load indication is calculated and the working bias is iteratively updated, and a fast reset is performed. The operating bias is restored to the stable point before reset, and the trap load indication is recalculated until the trap status flag is stable. The stable operating bias and trap status flag are then obtained.

6. The X-ray detection method based on perovskite materials as described in claim 5, characterized in that: The process involves acquiring electrical and optical channel signals, performing bias subtraction and response correction on the electrical channel signals according to an initial non-uniformity correction parameter table, adjusting the photoelectric channel fusion weights using trap status flags, and obtaining fused dose values ​​and pixel dose maps. The specific steps are as follows: The photocurrent signal of the electrical channel is collected to form an electrical signal matrix, and the fluorescence signal of the light-transmitting window is collected to form an optical signal matrix. The electrical signal matrix is ​​subjected to bias subtraction, gain correction and nonlinear compensation based on the initial non-uniformity correction parameter table to obtain the electrical channel correction signal; The electrical channel correction signal and the optical signal matrix are normalized to obtain pixel-level electrical channel dose estimates and optical channel dose estimates. The photoelectric fusion weights are calculated and weighted to obtain the fused dose value. The fused dose values ​​of all pixels are combined according to their spatial location to form a two-dimensional pixel dose map, thus obtaining the pixel dose map.

7. The X-ray detection method based on perovskite materials as described in claim 6, characterized in that: The initial non-uniformity correction parameter table is updated iteratively in small steps based on the fused dose value and pixel dose map to obtain the non-uniformity correction parameter table. The specific steps are as follows: Based on the fused dose value and pixel dose map, the target dose reference value is determined in a flat field or known dose distribution scene, and the residual of each pixel is calculated. When the trap status flag is stable, the pixel bias, gain coefficient and nonlinear correction parameter are iteratively updated in small steps using the residual of each pixel to obtain the parameter update amount. The update is paused when the trap status flag is that the trap is overfilled or underfilled. Based on the parameter update amount, spatial smoothing constraints are applied to the update amounts of adjacent pixels, and periodic cache accumulation is performed to obtain the non-uniformity correction parameter table.

8. The X-ray detection method based on perovskite materials as described in claim 7, characterized in that: The process involves calculating a comprehensive set of detection quality indicators by fusing dose values, non-uniformity correction parameter tables, and trap status flags to obtain the final dose map. The specific steps are as follows: Based on the fusion dose value, non-uniformity correction parameter table and trap status flag, flat field, oblique target and black field and low dose flat field data are collected sequentially to obtain flat field fusion dose map, oblique target fusion dose map and black field and low dose flat field sequence. Spatial non-uniformity index is calculated by flat-field fusion dose map, edge spread function and line spread function are extracted by oblique-side target fusion dose map to obtain MTF index, and noise equivalent dose index is obtained by statistically analyzing noise in black field and low-dose flat field sequences. Quality compliance is determined based on spatial non-uniformity index, MTF index, noise equivalent dose index, and trap status indicator. When the spatial non-uniformity index, MTF index, and noise equivalent dose index all meet the standards and the trap status indicator is stable, the final fusion dose map is obtained.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the X-ray detection method based on perovskite material as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the X-ray detection method based on perovskite materials as described in any one of claims 1 to 8.

Images