A quantum state high-fidelity discrimination method and device based on ion automatic positioning

Abstract

This invention discloses a high-fidelity quantum state discrimination method and device based on automatic ion localization. The invention constructs a dynamic closed-loop compensation system of ion position update, effective discrimination region reconstruction, and adaptive adjustment of discrimination parameters. This system addresses the degradation of discrimination fidelity over time due to imaging drift in multi-ion quantum state discrimination, achieving long-term stability and high-fidelity quantum state discrimination. When constructing the effective discrimination region for ions, a physical constraint model based on the ion trap potential field distribution is introduced, utilizing bright spot features and the physical property of ions aligning along the ion trap axis to form linear ion chains, establishing a correspondence between ions and bright spot regions. By extracting the feature parameters of the point spread function in real-time from the average bright state image, and adaptively constructing local pixel regions highly correlated with the actual fluorescence distribution of ions, random noise interference from background pixels is suppressed to the maximum extent, fundamentally improving the signal-to-noise ratio of multi-ion quantum state discrimination.

Description

Technical Field

[0001] This invention belongs to the field of trapped ion quantum information and quantum measurement technology, specifically relating to a high-fidelity quantum state discrimination method based on automatic ion localization, and also to a computer device. Background Technology

[0002] Trapped ion systems are one of the mature physical platforms for realizing quantum computing and quantum simulation. According to the DiVincenzo criterion, rapid and accurate readout of quantum states is an indispensable key element in quantum computing systems. In ion trap systems, electron shelving is typically used to probe the internal quantum states of ions: when the ion is in a bright state... When the ion is in the dark, it will produce stable fluorescence under the illumination of the detector light; When the fluorescence signal is detected, almost no fluorescence is produced, and quantum state discrimination can be achieved by detecting the fluorescence signal.

[0003] In practical experiments, common fluorescence detection methods include photomultiplier tubes (PMTs) and imaging-based EMCCD cameras. PMTs have high temporal resolution but are difficult to distinguish between multiple spatially adjacent ions. In contrast, EMCCD cameras can acquire spatial distribution information of multiple ions simultaneously, thus having a significant advantage in multi-bit ion trap systems.

[0004] However, existing methods for ion quantum state discrimination based on EMCCD images typically perform pixel statistical analysis directly on the entire image or a manually selected fixed area. These methods implicitly assume that the spatial position of ions on the EMCCD is stable over a long period of time. However, in actual systems, the position of ions on the imaging plane will slowly but not negligibly drift over time due to factors such as changes in ambient temperature, deformation of mechanical structures, and slight drifts in the optical path. When fixed pixel areas or manually selected areas are still used for statistical analysis, irrelevant background pixels or missing effective signal pixels are easily introduced, which significantly reduces the accuracy of quantum state discrimination, especially in the case of multiple ions.

[0005] Therefore, how to automatically and accurately acquire the real-time spatial position of ions on the imaging plane during quantum state discrimination, and then use only pixel information related to the actual fluorescence distribution of ions for statistical analysis, is a key technical problem in improving the readout fidelity of multi-ion quantum states. Summary of the Invention

[0006] The purpose of this invention is to address the aforementioned problems in the prior art by providing a high-fidelity quantum state discrimination method based on automatic ion localization, and also to provide a computer device.

[0007] The above-mentioned objectives of the present invention are achieved by the following technical means:

[0008] A high-fidelity quantum state discrimination method based on automatic ion localization includes the following steps:

[0009] Step 1: Process the acquired bright-state fluorescence images of multiple frames of ions into a bright-state average image;

[0010] Step 2: Process the acquired multi-frame dark-state fluorescence images of ions into a dark-state average image;

[0011] Step 3: Using the dark-state average image as the background noise distribution image, extract the bright spot regions from the bright-state average image;

[0012] Step 4: Filter out the regions where each ion is located in the bright spot region, and delineate the effective discrimination region of the corresponding ion in the bright spot region where each ion is located.

[0013] Step 5: Statistically analyze the probability distribution functions of bright state pixel intensity distribution and dark state pixel intensity distribution within the effective discrimination region of each ion, determine the discrimination parameters that maximize the distinction between bright and dark states, and discriminate the quantum states of each ion based on the discrimination parameters;

[0014] Bright pixels are the pixels of the average bright image, and dark pixels are the pixels of the average dark image;

[0015] Step 6: Set the update interval and update conditions. Every time the update interval expires or any update condition is met, repeat steps 1 to 5 to update the effective discrimination region, probability distribution function, discrimination parameters, and quantum state of the ion.

[0016] The specific regions where each ion is located, as described above, are obtained through the following steps:

[0017] Step 4.1: Calculate the pixel coordinates of the centroid of each bright spot region using image moments, and convert the pixel coordinates of each bright spot region into real coordinates;

[0018] Step 4.2: Sort the bright spot regions from largest to smallest based on their bright spot characteristics, and select the top-ranked regions. A bright spot area, The number of ions; then the obtained by screening The bright spot regions are arranged from largest to smallest along the ion trap axis according to the pixel coordinates of the centroid, so that the initial bright spot region and the pixel coordinates of the centroid of each ion in the linear ion chain are obtained respectively.

[0019] Step 4.3: Calculate the theoretical equilibrium position of each ion in the linear ion chain; then calculate the difference between the theoretical equilibrium position of each ion in the linear ion chain and the true coordinates of the centroid of the initially corresponding bright spot region. The difference between the theoretical equilibrium position of an ion and the true coordinates of the centroid of the initially corresponding bright spot region is less than or equal to a preset position deviation threshold. When the time is right, the bright spot region initially corresponding to the ion is determined to be the region where the ion is located.

[0020] Step 4.4: When the difference between the theoretical equilibrium position of the ion and the true coordinates of the centroid of the initially corresponding bright spot region is greater than the position deviation threshold... If the initial correspondence is incorrect, the bright spot region initially corresponding to the ion is not the region where the ion is located. Then, the difference between the theoretical equilibrium position of the ion with the incorrect initial correspondence and the true coordinates of the centroids of the remaining bright spot regions that are not the ion's region is calculated. The difference must be less than or equal to the position deviation threshold. The bright spot region corresponding to the true coordinates of the centroid is determined to be the region where the ion is located; finally, the bright spot region that is not the region where the ion is located is determined to be a false bright spot.

[0021] If, after step 4.4, the region where an ion is located is not identified, the current operation is stopped, and an ion dimming warning is issued to remind the user to reload the ion.

[0022] The effective discrimination region for the ions mentioned above is defined in the following way:

[0023] Using the pixel coordinates of the centroid of the bright spot region where each ion is located as the center, a local pixel region of a preset size is extracted from the bright state average image for each ion, and each local pixel region is the effective discrimination region for the corresponding ion.

[0024] In this context, the local pixel region of each ion is selected as a square or circular region covering the main lobe of the point spread function of the corresponding ion, wherein the square or circular region has the full width at half maximum (FWHM) of the point spread function as its side length or radius.

[0025] As mentioned above, the discrimination parameters include pixel intensity threshold. and effective pixel count threshold ;

[0026] The pixel intensity threshold Selected as Maximum ,in, Let be the probability distribution function of the intensity distribution of bright-state pixels within the effective discrimination region of ions. Let be the probability distribution function of the intensity distribution of dark-state pixels within the effective discrimination region of ions. The probability distribution function Medium pixel intensity Greater than the pixel intensity threshold The probability, The probability distribution function Medium pixel intensity Greater than the pixel intensity threshold The probability; the effective pixel count threshold The value is the average number of effective pixels within the effective discrimination region of the ion. Standard deviation of the number of effective pixels minus a preset multiple ;

[0027] Among them, effective pixels are those whose pixel intensity is greater than the pixel intensity threshold within the effective discrimination region of ions. Pixels;

[0028] Average number of effective pixels and standard deviation The following method was used: First, in each frame of the original bright fluorescence image, the pixel intensity greater than the pixel intensity threshold within the effective discrimination region of the ion was statistically analyzed. The number of pixels is the number of effective pixels; then according to The mean number of effective pixels is calculated from the number of effective pixels in the bright-state fluorescence image. and standard deviation , This represents the number of frames in the bright fluorescence image.

[0029] The quantum states of each ion are determined as described above in the following manner:

[0030] In the real-time acquired fluorescence images of ions, the number of effective pixels within the effective discrimination region of each ion is counted. When the number of effective pixels within the effective discrimination region of an ion is greater than or equal to the effective pixel count threshold, the ion is considered a valid ion. Then the ion is determined to be State; when the number of effective pixels within the effective discrimination region of an ion is less than the effective pixel count threshold. Then the ion is determined to be The quantum state of all ions is determined by distinguishing them.

[0031] The update conditions mentioned above include:

[0032] Update condition 1: The center offset of the effective discrimination region of ions relative to the effective discrimination region of ions defined in the previous update exceeds a preset threshold;

[0033] Update condition 2: The mean pixel intensity of all bright pixels in the effective discrimination region of ions in the bright average image deviates from the set percentage of the mean pixel intensity of bright pixels in the effective discrimination region of ions in the previously updated bright average image.

[0034] Update condition 3: The fidelity obtained in the calibration experiment is lower than the preset allowable range.

[0035] The theoretical equilibrium positions of the ions, as described above, are calculated in the following manner:

[0036] Filtered The largest true coordinate among the true coordinates of the centroids of the bright spot regions and the corresponding bright spot region are the reference position and the reference bright spot region, respectively. The theoretical equilibrium position of the ion initially corresponding to the bright spot region with the reference position as the reference is the first... Theoretical equilibrium position of each ion Calculated based on the following formula:

[0037] ;

[0038] In the formula, This represents the total potential energy of the linear ionic chain. is the axial secular frequency of the ion trap; It refers to the ion mass; It is the vacuum permittivity; It is the elementary charge; and All are ion serial numbers. and The values ​​are all ; For the first The theoretical equilibrium positions of the ions, For the first The theoretical equilibrium positions of the ions, for The absolute value; The total potential energy of the linear ionic chain with respect to the first Theoretical equilibrium position of each ion The partial derivatives of .

[0039] The average bright-state image, as described above, is obtained in the following way:

[0040] First, the multi-frame bright fluorescence images are averaged pixel by pixel to obtain the average pixel intensity of each bright pixel. The initial bright average image is then constructed using the average pixel intensity of all bright pixels.

[0041] Secondly, the initial bright state average image is smoothed to obtain the bright state average image;

[0042] The dark state average image is obtained by performing pixel-by-pixel averaging on multiple frames of dark state fluorescence images to obtain the average pixel intensity of each dark state pixel, and constructing the dark state average image based on the average pixel intensity of all dark state pixels.

[0043] The bright spot regions are extracted in the following way: a background noise distribution image is constructed using the pixel intensity of each dark pixel in the dark average image as the background noise value of the corresponding pixel; pixels in the bright average image whose difference between the average pixel intensity and the background noise value is greater than a preset segmentation threshold are selected; and each bright spot region is constructed in the bright average image using the selected pixels.

[0044] A computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of a quantum state high-fidelity discrimination method based on automatic ion localization as described above.

[0045] Compared with the prior art, the present invention has the following advantages:

[0046] (1) Constructing a dynamic closed-loop mechanism to ensure system stability: This invention establishes a dynamic closed-loop compensation system of ion position update - effective discrimination region reconstruction - discrimination parameter adaptive adjustment through periodic or condition-triggered relocation and parameter adaptive update. It can identify and automatically correct imaging shifts caused by environmental temperature fluctuations, mechanical structure deformation or optical path drift in real time, eliminate statistical mismatch caused by physical position drift in existing methods, significantly improve the long-term operational stability of quantum state readout, solve the defect of discrimination fidelity degrading over time due to imaging drift in multi-ion quantum state discrimination, and achieve long-term stable and high-fidelity quantum state discrimination.

[0047] (2) Based on optical and physical characteristics, the present invention accurately selects the region to maximize the signal-to-noise ratio: The present invention abandons the existing method of manually presetting fixed pixel regions, utilizes the bright spot characteristics of the bright spot region and the physical characteristics of ions arranged sequentially in the ion trap axis to form a linear ion chain, introduces a physical constraint model based on the ion trap potential field distribution, establishes the correspondence between ions and bright spot regions, eliminates false bright spots, and identifies the situation of ion darkening.

[0048] Furthermore, by utilizing the bright-state average image to extract the feature parameters of the point spread function of each ion and the pixel coordinates of the centroid in real time, and by adaptively constructing a local pixel region that is highly correlated with the actual fluorescence distribution of the ions, the random noise interference of the background pixels is suppressed to the maximum extent, fundamentally improving the signal-to-noise ratio of multi-ion quantum state discrimination. Attached Figure Description

[0049] Figure 1 This is a flowchart illustrating the method of the present invention. Detailed Implementation

[0050] To facilitate understanding and implementation of the present invention by those skilled in the art, the present invention will be further described in detail below with reference to embodiments. The embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0051] Example 1:

[0052] like Figure 1 As shown, a high-fidelity quantum state discrimination method based on automatic ion localization includes the following steps:

[0053] Step 1: Acquire multiple frames of bright-state fluorescence images of ions, and process them pixel by pixel to obtain a bright-state average image. This includes the following steps:

[0054] Step 1.1, Ion Bright-State Fluorescence Image Acquisition: In an ion trap containing two or more ions, prepare all ions within the ion trap to... (that is, all ions are prepared simultaneously to) The fluorescence of ions is imaged and detected using an EMCCD (electron multiplier charge-coupled device) camera, acquiring multiple frames of bright-state fluorescence images.

[0055] Step 1.2: Generate the initial bright state average image: Perform pixel-by-pixel averaging on the multiple frames of bright state fluorescence images to obtain the average pixel intensity of each bright state pixel, and generate the initial bright state average image based on the average pixel intensity of each bright state pixel.

[0056] Step 1.3: To suppress the influence of high-frequency noise on bright spot recognition, Gaussian blur is used to smooth the initial bright state average image to obtain the bright state average image.

[0057] Step 2: Acquire multiple frames of dark-state fluorescence images of ions, and process them pixel-by-pixel to obtain a dark-state average image. This specifically includes the following steps:

[0058] Step 2.1: Prepare all ions in the ion trap to... (that is, all ions are prepared simultaneously to) The state is detected and multiple frames of dark-state fluorescence images of ions are acquired during the state detection process.

[0059] Step 2.2: Perform pixel-by-pixel averaging on the multi-frame dark fluorescence images to obtain the average pixel intensity of each dark pixel. Generate a dark average image using the average pixel intensity of each dark pixel. Use the spatial distribution of the average pixel intensity of the dark pixels as the background noise distribution for subsequent image processing and determination of discrimination parameters.

[0060] Among them, bright state pixels are pixels in bright state fluorescence images and bright state average images, and dark state pixels are pixels in dark state fluorescence images and dark state average images. By acquiring multiple frames of images and performing pixel-by-pixel averaging, shot noise and random background noise can be effectively suppressed, making the spatial distribution of bright spots of ion fluorescence more stable and clear.

[0061] Step 3: Using the dark average image as the background noise distribution image, extract bright spot regions from the bright average image. Specifically, construct a background noise distribution image using the pixel intensity of each dark pixel in the dark average image as the background noise value of the corresponding pixel. Set a segmentation threshold and filter out pixels in the bright average image whose difference between the average pixel intensity and the background noise value is greater than the segmentation threshold. Construct each bright spot region using the filtered pixels, thereby separating the bright spot region from the background.

[0062] As one possible implementation, the segmentation threshold can be selected from the global brightness standard deviation of the bright-state average image or a preset signal-to-noise ratio multiple.

[0063] Step 4: Filter out the regions containing each ion within the bright spot area, and delineate the effective discrimination region for each ion within the bright spot area containing each ion. This includes the following steps:

[0064] Step 4.1 First, for each bright spot region obtained by segmentation, the pixel coordinates of the centroid of each bright spot region are calculated using image moments, and the pixel coordinates of the centroid are used as the spatial position of the corresponding ion on the EMCCD imaging plane.

[0065] Then, the pixel coordinates of the centroid of the bright spot region on the EMCCD imaging plane are converted into true coordinates (e.g., the true distance represented by one pixel spacing in this laboratory is 0.6 μm (micrometers)) so as to calculate the difference with the theoretical equilibrium position of the ions obtained in subsequent steps.

[0066] Step 4.2: Sort the bright spot regions from largest to smallest based on their bright spot characteristics, and select the top-ranked regions. A bright spot area, For the number of ions, the bright spot characteristic can be selected as bright spot intensity or bright spot area;

[0067] Utilizing the physical property that ions arrange themselves sequentially along the axial direction of the ion trap to form a linear ion chain, the selected ions are then... The bright spot regions are arranged from largest to smallest along the ion trap axis according to the pixel coordinates of their centroids, thereby establishing the initial correspondence between each ion in the linear ion chain and each bright spot region, and obtaining the initial bright spot region and the pixel coordinates of the centroid of each ion in the linear ion chain.

[0068] Step 4.3: Introduce a physical constraint model based on the potential field distribution of the ion trap, and calculate the theoretical equilibrium positions of each ion in the linear ion chain, specifically:

[0069] Total potential energy of linear ionic chains The expression is:

[0070] (1);

[0071] This embodiment uses the results obtained through screening. The largest true coordinate among the true coordinates of the centroid of the bright spot region and the corresponding bright spot region are taken as the reference position and the reference bright spot region, respectively. Taking the reference position as the theoretical equilibrium position of the ion initially corresponding to the reference bright spot region, the first ion in the linear ion chain... Theoretical equilibrium position of each ion Calculated based on the following formula:

[0072] (2);

[0073] In the formula, It refers to the ion mass; This is the axial secular frequency; It is the vacuum permittivity; It is the elementary charge; and All are ion serial numbers. and The values ​​are all , The number of ions; For the first The theoretical equilibrium positions of the ions, For the first The theoretical equilibrium positions of the ions, for The absolute value; The total potential energy of the linear ionic chain with respect to the first Theoretical equilibrium position of each ion The partial derivatives of .

[0074] Next, the theoretical equilibrium positions of each ion in the linear ion chain are calculated to differ from the actual coordinates of the centroid of the corresponding initial bright spot region; a position deviation threshold is then set. When the difference between the theoretical equilibrium position of the ion and the true coordinates of the centroid of the initial bright spot region corresponding to the ion is less than or equal to the position deviation threshold. When the time is right, the bright spot region initially corresponding to the ion is determined to be the region where the ion is located.

[0075] In this embodiment, the position deviation threshold It is selected as half of the minimum inter-ion spacing.

[0076] Step 4.4: When the difference between the theoretical equilibrium position of the ion and the true coordinates of the centroid of the initially corresponding bright spot region is greater than the position deviation threshold... If the initial correspondence is incorrect, the bright spot region initially corresponding to the ion is not the region where the ion is located.

[0077] Next, calculate the difference between the theoretical equilibrium positions of the ions with incorrect initial correspondences and the true coordinates of the centroids of the bright spot regions that are not in the ion's region, and set the difference to be less than the position deviation threshold. The bright spot region corresponding to the true coordinates of the centroid is determined to be the region where the ions are located.

[0078] Finally, bright spots that are not located in the ion region are identified as false bright spots and removed.

[0079] Step 4.5: If, after step 4.4, the region where an ion is located is not determined, the current quantum logic operation is stopped, and an ion dimming warning is issued, reminding the user that ion reloading is required.

[0080] Step 4.6: Using the pixel coordinates of the centroid of the bright spot region where each ion is located as the center, extract a local pixel region of a predetermined size for each ion in the bright state average image, and use each local pixel region as the effective discrimination region for the corresponding ion.

[0081] As one possible implementation, the local pixel region of a predetermined size can be selected in the following way: the local pixel region is determined by the point spread function of ion fluorescence in the imaging system. In this embodiment, it is selected as a square or circular region covering the main lobe of the point spread function, so that the selected region can cover most of the energy of the ion fluorescence signal, while avoiding the introduction of too many background pixels. The square or circular region covering the main lobe of the point spread function has the full width at half maximum (FWHM) of the point spread function as its side length or radius, and the center of the square or circular region is the pixel coordinate of the centroid.

[0082] In defining the effective discrimination region of ions, this invention combines optical and physical properties to accurately select the region, thereby maximizing the signal-to-noise ratio. It abandons the existing method of manually presetting fixed pixel regions and utilizes the bright spot characteristics of the bright spot region and the physical property of ions arranging in sequence along the ion trap axis to form a linear ion chain to establish an initial correspondence between ions and bright spot regions. Then, it introduces a physical constraint model based on the potential field distribution of the ion trap to eliminate false bright spots and ion darkening, and finally identifies the bright spot region where the ions are actually located.

[0083] Furthermore, based on the real-time extraction of the feature parameters of the point spread function (full width at half maximum in this embodiment) and the pixel coordinates of the centroid of the bright state average image, the size of the local pixel region used for discrimination is dynamically and adaptively limited. Compared with directly performing statistics on the entire image or a fixed selected area, the effective discrimination region of the method of the present invention can effectively cover the main spatial distribution of ion fluorescence, suppress the random noise interference of irrelevant background pixels to the maximum extent, and fundamentally improve the signal-to-noise ratio of multi-ion quantum state discrimination.

[0084] Step 5: Statistically calculate the probability distribution functions of bright-state pixel intensity distribution and dark-state pixel intensity distribution within the effective discrimination region of each ion, determine the discrimination parameters that maximize the distinction between bright and dark states, and then discriminate the quantum states of each ion based on these parameters. This process includes the following steps:

[0085] As one possible implementation, for each ion's effective discrimination region, based on the probability distribution function of the bright state pixel intensity distribution and the probability distribution function of the dark state pixel intensity distribution within the effective discrimination region of each ion, the pixel intensity threshold and the effective pixel number threshold that can maximize the distinction between the bright state and the dark state are automatically determined as discrimination parameters for quantum state discrimination. The discrimination parameters can be automatically selected based on statistical characteristics and do not depend on manual experience.

[0086] Step 5.1: Let the probability distribution function of the bright pixel intensity distribution within the effective discrimination region of the ion be . The probability distribution function of the dark pixel intensity distribution within the effective discrimination region of ions is: ;

[0087] Then pixel intensity threshold Selected as Maximum , The probability distribution function Medium pixel intensity Greater than the pixel intensity threshold The probability, The probability distribution function Medium pixel intensity Greater than the pixel intensity threshold The probability of.

[0088] Effective pixel count threshold Pixel intensity can exceed within the effective ion discrimination region The effective pixel count distribution is determined, and in this embodiment, the mean number of effective pixels within the effective discrimination region of ions is... Standard deviation of the number of effective pixels minus a preset multiple As a discrimination threshold, it aims to minimize the probability of misclassification of bright and dark states in a statistical sense;

[0089] Among them, the average number of effective pixels and standard deviation The following method was used: First, in each frame of the original bright fluorescence image, the pixel intensity greater than the pixel intensity threshold within the effective discrimination region of the ion was statistically analyzed. The number of pixels is the number of effective pixels, where the first... The effective pixel count of the bright-state fluorescence image is , This represents the frame number of the bright fluorescence image. , The number of frames in the bright fluorescence image; then according to The mean number of effective pixels is calculated from the number of effective pixels in the bright-state fluorescence image. and standard deviation .

[0090] This invention introduces a pixel intensity threshold in the process of determining ion quantum states. and effective pixel count threshold The dual-threshold decision optimization has higher robustness than the simple intensity centroid method.

[0091] Step 5.2: Determine the quantum state of each ion: In the real-time acquired ion fluorescence image, count the number of effective pixels in the effective discrimination region of each ion. If the number of effective pixels in the effective discrimination region of an ion is greater than or equal to the effective pixel count threshold... Then the ion is determined to be State; if the number of effective pixels within the effective discrimination region of an ion is less than the effective pixel count threshold. Then the ion is determined to be state.

[0092] Step 5.3: Perform step 5.2 within the effective discrimination region of each ion to discriminate the quantum states of all ions, thereby obtaining the quantum states of all ions.

[0093] As one possible implementation, a likelihood function for both bright and dark states can be constructed. Maximum likelihood estimation can then be used to determine the first likelihood value of the pixel intensity distribution within the effective discrimination region of each ion in the real-time acquired fluorescence image, which corresponds to the probability distribution function of the bright state pixel intensity distribution, and the second likelihood value of the pixel intensity distribution within the dark state pixel intensity distribution. If the first likelihood value is greater than the second likelihood value, then the ion is considered to be... If the first likelihood value is less than the second likelihood value, then the ion is... state.

[0094] Step 6: Set the update interval and update conditions. Each time the update interval expires or any update condition is met, repeat steps 1-5 to update the effective discrimination region, probability distribution function, discrimination parameters, and quantum state of the ion. This compensates for system drift and ensures the long-term stability of quantum state discrimination. Specifically, this includes the following process:

[0095] The update conditions in this embodiment include:

[0096] Update condition 1: The center offset of the effective discrimination region of the ion spatial position relative to the effective discrimination region of the ion defined in the last update exceeds a preset threshold. Specifically, the preset threshold for the center offset is 20% of the full width at half maximum of the point spread function or 10% of the ion spacing. Once the offset exceeds this value, it is considered that the imaging drift has affected the readout fidelity.

[0097] Update condition 2: The mean pixel intensity of all bright pixels within the effective discrimination region of ions in the bright-state average image deviates from the mean pixel intensity of bright pixels within the effective discrimination region of ions in the previously updated bright-state average image by more than 15%.

[0098] In quantum computing, in order to monitor readout fidelity, a predetermined number of controlled experiments, namely calibration experiments, are usually inserted to prepare and measure known states. Calibration experiments refer to the process of preparing known quantum states and using current parameters to make judgments in order to evaluate the accuracy of the judgments.

[0099] Update condition 3: The fidelity obtained in the calibration experiment is less than 99%.

[0100] This invention determines the quantum state of ions with high fidelity by setting update time intervals and update conditions. Each update is an adaptive adjustment of the decision parameters: each update compares the bright spot region where the ion is located and the point spread function parameters (e.g., full width at half maximum or energy distribution radius) of the bright spot region in the average bright state image at different times compared to the previous update, which can determine whether the distribution of ion fluorescence in the local pixel region has changed. When a change in ion position or point spread function is detected, the pixel intensity distribution and the number of effective pixels will change accordingly, thereby updating the decision parameters. Automatic or conditional updates can further suppress systematic biases and improve the long-term stability and robustness of quantum state discrimination.

[0101] To address the systematic errors caused by bright spot drift and discrimination region mismatch in statistical decision-making under multi-ion conditions, which are not resolved in existing technologies, this invention no longer assumes that the position of ions on the EMCCD imaging plane remains constant. Instead, it achieves automatic compensation for imaging drift and environmental disturbances through a closed-loop processing flow of ion position update—effective discrimination region reconstruction—adaptive adjustment of discrimination parameters, thereby significantly improving the accuracy and long-term stability of multi-ion quantum state discrimination. Under the same exposure time and number of ions, the quantum state discrimination accuracy of existing methods shows a significant downward trend over experimental time. However, the method of this invention can improve the multi-ion bright and dark state discrimination accuracy from about 95% to over 99.9% and maintain a stable discrimination accuracy during long-term operation. This invention can significantly reduce the discrimination error caused by ion position drift, verifying its significant beneficial effect.

[0102] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0103] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0104] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0105] It should be noted that the embodiments described in this invention are merely illustrative of the spirit of the invention. Those skilled in the art to which this invention pertains can make various modifications or additions to the described embodiments or use similar methods to substitute them, without departing from the spirit of the invention or exceeding the scope defined by the appended claims.

Claims

1. A high-fidelity quantum state discrimination method based on automatic ion localization, characterized in that, Includes the following steps: Step 1: Process the acquired bright-state fluorescence images of multiple frames of ions into a bright-state average image; Step 2: Process the acquired multi-frame dark-state fluorescence images of ions into a dark-state average image; Step 3: Using the dark-state average image as the background noise distribution image, extract the bright spot regions from the bright-state average image; Step 4: Filter out the regions where each ion is located in the bright spot region, and delineate the effective discrimination region of the corresponding ion in the bright spot region where each ion is located. Step 5: Statistically analyze the probability distribution functions of bright state pixel intensity distribution and dark state pixel intensity distribution within the effective discrimination region of each ion, determine the discrimination parameters that maximize the distinction between bright and dark states, and discriminate the quantum states of each ion based on the discrimination parameters; Bright pixels are the pixels of the average bright image, and dark pixels are the pixels of the average dark image; Step 6: Set the update interval and update conditions. Every time the update interval expires or any update condition is met, repeat steps 1 to 5 to update the effective discrimination region, probability distribution function, discrimination parameters, and quantum state of the ion.

2. The high-fidelity quantum state discrimination method based on automatic ion localization according to claim 1, characterized in that, The specific ion regions of each ion are obtained through the following steps: Step 4.1: Calculate the pixel coordinates of the centroid of each bright spot region using image moments, and convert the pixel coordinates of each bright spot region into real coordinates; Step 4.2: Sort the bright spot regions from largest to smallest based on their bright spot characteristics, and select the top-ranked regions. A bright spot area, The number of ions; then the obtained by screening The bright spot regions are arranged from largest to smallest along the ion trap axis according to the pixel coordinates of the centroid, so that the initial bright spot region and the pixel coordinates of the centroid of each ion in the linear ion chain are obtained respectively. Step 4.3: Calculate the theoretical equilibrium position of each ion in the linear ionic chain; Next, the difference between the theoretical equilibrium position of each ion in the linear ion chain and the true coordinates of the centroid of the initially corresponding bright spot region is calculated. The position is considered positive when the difference between the theoretical equilibrium position of the ion and the true coordinates of the centroid of the initially corresponding bright spot region is less than or equal to a preset position deviation threshold. When the time is right, the bright spot region initially corresponding to the ion is determined to be the region where the ion is located. Step 4.4: When the difference between the theoretical equilibrium position of the ion and the true coordinates of the centroid of the initially corresponding bright spot region is greater than the position deviation threshold... If the initial correspondence is incorrect, the bright spot region initially corresponding to the ion is not the region where the ion is located. Next, calculate the difference between the theoretical equilibrium position of the ion with the initial incorrect correspondence and the true coordinates of the centroid of the bright spot region that is not the ion's location. Set the difference to be less than or equal to the position deviation threshold. The bright spot region corresponding to the true coordinates of the centroid is determined to be the region where the ion is located; finally, the bright spot region that is not the region where the ion is located is determined to be a false bright spot.

3. The high-fidelity quantum state discrimination method based on automatic ion localization according to claim 2, characterized in that, If, after step 4.4, the region where an ion is located is not identified, the current operation is stopped, and an ion dimming warning is issued to remind the user to reload the ion.

4. The high-fidelity quantum state discrimination method based on automatic ion localization according to claim 3, characterized in that, The effective discrimination region of the ions is defined in the following manner: Using the pixel coordinates of the centroid of the bright spot region where each ion is located as the center, a local pixel region of a preset size is extracted from the bright state average image for each ion, and each local pixel region is the effective discrimination region for the corresponding ion. In this context, the local pixel region of each ion is selected as a square or circular region covering the main lobe of the point spread function of the corresponding ion, wherein the square or circular region has the full width at half maximum (FWHM) of the point spread function as its side length or radius.

5. The high-fidelity quantum state discrimination method based on automatic ion localization according to claim 4, characterized in that, The discrimination parameters include pixel intensity threshold. and effective pixel count threshold ; The pixel intensity threshold Selected as Maximum ,in, Let be the probability distribution function of the intensity distribution of bright-state pixels within the effective discrimination region of ions. Let be the probability distribution function of the intensity distribution of dark-state pixels within the effective discrimination region of ions. The probability distribution function Medium pixel intensity Greater than the pixel intensity threshold The probability, The probability distribution function Medium pixel intensity Greater than the pixel intensity threshold The probability; the effective pixel count threshold The value is the average number of effective pixels within the effective discrimination region of the ion. Standard deviation of the number of effective pixels minus a preset multiple ; Among them, effective pixels are those whose pixel intensity is greater than the pixel intensity threshold within the effective discrimination region of ions. Pixels; Average number of effective pixels and standard deviation The following method was used: First, in each frame of the original bright fluorescence image, the pixel intensity greater than the pixel intensity threshold within the effective discrimination region of the ion was statistically analyzed. The number of pixels is the number of effective pixels; then according to The mean number of effective pixels is calculated from the number of effective pixels in the bright-state fluorescence image. and standard deviation , This represents the number of frames in the bright fluorescence image.

6. The high-fidelity quantum state discrimination method based on automatic ion localization according to claim 5, characterized in that, The quantum states of each ion are determined in the following way: In the real-time acquired fluorescence images of ions, the number of effective pixels within the effective discrimination region of each ion is counted. When the number of effective pixels within the effective discrimination region of an ion is greater than or equal to the effective pixel count threshold, the ion is considered a valid ion. Then the ion is determined to be State; when the number of effective pixels within the effective discrimination region of an ion is less than the effective pixel count threshold. Then the ion is determined to be The quantum state of all ions is determined by distinguishing them.

7. The high-fidelity quantum state discrimination method based on automatic ion localization according to claim 6, characterized in that, The update conditions include: Update condition 1: The center offset of the effective discrimination region of ions relative to the effective discrimination region of ions defined in the previous update exceeds a preset threshold; Update condition 2: The mean pixel intensity of all bright pixels in the effective discrimination region of ions in the bright average image deviates from the set percentage of the mean pixel intensity of bright pixels in the effective discrimination region of ions in the previously updated bright average image. Update condition 3: The fidelity obtained in the calibration experiment is lower than the preset allowable range.

8. The high-fidelity quantum state discrimination method based on automatic ion localization according to claim 2, characterized in that, The theoretical equilibrium position of the ion is calculated in the following way: Filtered The largest true coordinate among the true coordinates of the centroids of the bright spot regions and the corresponding bright spot region are the reference position and the reference bright spot region, respectively. The theoretical equilibrium position of the ion initially corresponding to the bright spot region with the reference position as the reference is the first... Theoretical equilibrium position of each ion Calculated based on the following formula: ； In the formula, This represents the total potential energy of the linear ionic chain. is the axial secular frequency of the ion trap; It refers to the ion mass; It is the vacuum permittivity; It is the elementary charge; and All are ion serial numbers. and The values ​​are all ; For the first The theoretical equilibrium positions of the ions, For the first The theoretical equilibrium positions of the ions, for The absolute value; The total potential energy of the linear ionic chain with respect to the first Theoretical equilibrium position of each ion The partial derivatives of .

9. The high-fidelity quantum state discrimination method based on automatic ion localization according to claim 1, characterized in that, The average bright-state image is obtained in the following way: First, the multi-frame bright fluorescence images are averaged pixel by pixel to obtain the average pixel intensity of each bright pixel. The initial bright average image is then constructed using the average pixel intensity of all bright pixels. Secondly, the initial bright state average image is smoothed to obtain the bright state average image; The dark state average image is obtained by performing pixel-by-pixel averaging on multiple frames of dark state fluorescence images to obtain the average pixel intensity of each dark state pixel, and constructing the dark state average image based on the average pixel intensity of all dark state pixels. The bright spot regions are extracted in the following way: a background noise distribution image is constructed using the pixel intensity of each dark pixel in the dark average image as the background noise value of the corresponding pixel; pixels in the bright average image whose difference between the average pixel intensity and the background noise value is greater than a preset segmentation threshold are selected; and each bright spot region is constructed in the bright average image using the selected pixels.

10. 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 quantum state high-fidelity discrimination method based on automatic ion localization as described in any one of claims 1 to 9.

Images