Method for improving optical precision and optimizing image calibration error of miniature microscopic system

Abstract

The invention relates to the field of microscopic imaging, and discloses a miniature microscopic system optical precision improvement and image calibration error optimization method, which comprises the following steps: establishing an optical distortion model and a mechanical disturbance model, and respectively describing a static error and a dynamic error; constructing a joint dynamic error model, and uniformly expressing a coupling relationship between optical distortion and mechanical disturbance; decoupling and separating the joint error through a space-time separation method to obtain a static error component and a dynamic error component; performing iterative optimization on the optical distortion parameter and the mechanical disturbance parameter based on an optimization algorithm to minimize a system calibration error; a real-time sensor is utilized to obtain dynamic disturbance data, and real-time dynamic compensation is performed on an image calibration error in combination with an optimization result, so that a high-precision calibration result is obtained. The method can effectively improve the calibration precision of the microscopic system in a dynamic environment, remarkably improves the imaging quality of the system, and has a wide application prospect.

Description

Technical Field

[0001] The present invention relates to the field of microscopic imaging technology, and in particular to a method for improving the optical accuracy of a micro-microscopic system and optimizing image calibration errors. Background Art

[0002] Microscope systems are widely used in fields such as biomedicine, materials science, and industrial inspection, and have extremely high requirements for imaging accuracy and stability. However, in practical applications, since microscope systems usually work in complex dynamic environments, such as mechanical vibrations, thermal expansion or tilt errors during equipment operation, and the distortion characteristics of the optical devices themselves, the calibration accuracy and imaging quality of the system are easily seriously affected. Most of the existing microscope system calibration methods are based on the optimization of static environments, mainly for parameter calibration or simple geometric correction of optical distortion, but these methods are usually unable to effectively deal with the complex coupling of optical and mechanical errors in dynamic environments.

[0003] In a dynamic environment, the microscopic system is not only affected by the static error caused by optical distortion, but also by the dynamic error caused by mechanical disturbance. The dynamic error is usually time-varying and uncertain, which further increases the complexity of system calibration. In the existing technology, optical error and mechanical error are usually analyzed separately in error processing, but there is a lack of unified modeling and optimization strategies, which makes it impossible to accurately separate the contributions of different error sources, and the calibration effect is often difficult to meet the requirements of high-precision microscopic imaging. In addition, for real-time compensation, the compensation methods in the existing technology lack the ability to adapt to dynamic disturbances, the compensation accuracy is limited, and the real-time performance and stability in complex scenes are difficult to guarantee.

[0004] Therefore, current technology has not yet been able to achieve unified modeling, precise separation, and efficient compensation of optical distortion and dynamic mechanical disturbances. In particular, in terms of system error calibration in dynamic environments, there are problems of insufficient accuracy, poor adaptability, and low real-time performance. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a method for improving the optical accuracy of a micro-microscope system and optimizing image calibration errors, which solves the problem that the calibration errors caused by optical distortion and mechanical disturbances in a microscopic system under a dynamic environment cannot be accurately separated and compensated in real time.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: A method for improving the optical accuracy of a micro-microscope system and optimizing image calibration errors, comprising the following steps:

[0007] Establish an optical distortion model and a mechanical disturbance model to describe the static error caused by optical distortion and the dynamic error caused by mechanical disturbance respectively;

[0008] Construct a joint dynamic error model of optical distortion and mechanical disturbance;

[0009] Decouple and separate the joint dynamic error, and separate the dynamic error into static and dynamic components;

[0010] Based on the decoupled error model, the optical distortion parameters and mechanical disturbance parameters are optimized to minimize the system calibration error.

[0011] Dynamic disturbance data is acquired through real-time sensors, and image calibration errors are compensated in real time based on the optimization results to obtain calibrated high-precision images.

[0012] Preferably, the optical distortion model includes radial distortion and tangential distortion, the optical distortion error is determined by multiple distortion parameters, including at least radial distortion parameters and tangential distortion parameters, and the radial distortion error increases nonlinearly with increasing optical center distance.

[0013] Preferably, the mechanical disturbance model is used to describe the displacement and tilt errors of the system in a dynamic environment, including:

[0014] Dynamic displacement error, which is caused by external vibration and manifests itself as amplitude changes along the optical axis and dynamic fluctuations of vibration frequency;

[0015] Tilt angle error, which is caused by the tilt of mechanical components, manifests itself as an angular deviation of the optical axis of the optical system during the imaging process, resulting in a deviation in the optical path of the imaging area.

[0016] Preferably, the step of decoupling and separating the joint dynamic error comprises:

[0017] Decompose the dynamic error into static error and dynamic error by using time-space decomposition;

[0018] Frequency domain analysis is used to extract dynamic error components through high-pass filtering and static error components through low-pass filtering.

[0019] Preferably, in the decoupling process of the dynamic error, the accuracy of error separation is improved by fusion analysis of multidimensional sensor data of the mechanical disturbance signal, wherein the multidimensional sensor data includes time series information of acceleration, angular velocity and displacement, and the fusion method adopts weighted filtering or Kalman filtering method.

[0020] Preferably, the step of optimizing the optical distortion parameters and the mechanical disturbance parameters and minimizing the system calibration error is implemented based on minimizing the objective function, the objective function is the minimum value of the sum of the squares of the dynamic error and the static error, and the optimization constraints include:

[0021] Static error continuity constraint requires the static error model to be second-order continuously differentiable;

[0022] The dynamic error amplitude constraint requires that the dynamic error component does not exceed a preset threshold.

[0023] Preferably, the error optimization process is solved by a gradient descent method, comprising the following steps:

[0024] Initialize optical distortion parameters and mechanical perturbation parameters;

[0025] Calculate the total error of the current system;

[0026] Calculate the gradient of each parameter according to the error function;

[0027] The parameters are iteratively updated based on the gradient until a preset convergence condition is reached.

[0028] Preferably, the step of performing real-time dynamic compensation for the image calibration error is based on dynamic disturbance data acquired by the sensor and includes:

[0029] Acquire the displacement and tilt data of mechanical disturbance in real time through acceleration sensor and gyroscope;

[0030] Dynamically calibrate and compensate image coordinates according to mechanical disturbance data;

[0031] The compensated image coordinates are input into the system for high-precision image reconstruction.

[0032] Preferably, the compensation model of the real-time dynamic compensation dynamically adjusts the compensation coefficient according to the frequency and amplitude of the dynamic disturbance to eliminate the influence of high-frequency vibration on the image.

[0033] The present invention also provides a system for improving the optical accuracy of a micro-microscope system and optimizing image calibration errors, comprising:

[0034] Optical distortion modeling module, used to build models describing radial distortion and tangential distortion;

[0035] Mechanical disturbance modeling module, used to build displacement and tilt models to describe dynamic disturbances;

[0036] Error decoupling module, used for time-space decomposition and frequency-domain analysis of joint dynamic errors;

[0037] An optimization module, used to optimize distortion parameters and perturbation parameters based on objective functions and constraints;

[0038] A real-time compensation module is used to compensate and calibrate dynamic disturbances based on real-time sensor data;

[0039] The data acquisition module is used to obtain dynamic disturbance signals and real-time data of the optical system.

[0040] The present invention provides a method for improving the optical accuracy of a micro-microscope system and optimizing image calibration errors. It has the following beneficial effects:

[0041] 1. The present invention achieves accurate separation and correction of microscopic system errors in dynamic environments by constructing a joint dynamic error model and optimizing optical distortion and mechanical disturbance parameters. The optimized parameters can effectively reduce the comprehensive calibration error of the system and ensure that the calibration accuracy reaches the sub-pixel level.

[0042] 2. The present invention obtains dynamic disturbance data and performs compensation through real-time sensors, and can quickly respond to the impact of dynamic factors such as mechanical vibration and tilt on the system. Its real-time dynamic compensation function ensures the stability and reliability of the microscopic system under complex dynamic conditions.

[0043] 3. Based on the correction of optical distortion and dynamic error, the present invention realizes high-precision image calibration, significantly reducing image distortion and blurring caused by system errors. The imaging quality of the optimized microscopic system is greatly improved, meeting the requirements of precision measurement and high-resolution imaging.

[0044] 4. The present invention significantly accelerates the parameter optimization and compensation process by combining the gradient optimization algorithm and real-time filtering technology. Its efficient error handling mechanism shortens the calibration time, reduces the dependence of system operation on environmental stability, and improves the overall work efficiency of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a schematic diagram of the method flow of the present invention;

[0046] Figure 2 It is a schematic diagram of the system structure of the present invention.

[0047] Among them, 10, optical distortion modeling module; 20, mechanical disturbance modeling module; 30, error decoupling module; 40, optimization module; 50, real-time compensation module; 60, data acquisition module. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the specification of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] Please refer to the attached Figure 1The present invention provides a method for improving the optical accuracy of a micro-microscope system and optimizing image calibration errors. The method establishes optical distortion and mechanical disturbance models, constructs a joint error model, decouples dynamic errors, optimizes distortion parameters and disturbance parameters, and combines real-time sensor feedback to achieve dynamic calibration compensation, thereby significantly improving the optical accuracy of the micro-microscope system.

[0050] The method for improving the optical accuracy of a micro-microscope system and optimizing image calibration errors may include the following steps:

[0051] S1. Establish optical distortion model and mechanical disturbance model;

[0052] S2, construct a joint dynamic error model of optical distortion and mechanical disturbance;

[0053] S3, decoupling and separating the joint dynamic error;

[0054] S4, optimizing optical distortion parameters and mechanical disturbance parameters;

[0055] S5. Real-time dynamic error compensation to obtain calibrated high-precision images.

[0056] Each step of the method of the present invention is described in detail below.

[0057] For step S1, in this embodiment, an optical distortion model and a mechanical disturbance model are first established to describe the static error caused by optical distortion and the dynamic error caused by mechanical disturbance, respectively. This step is the basis of the entire optimization process, and its core is to accurately model and describe the main sources of system errors to ensure that the error decoupling, optimization and compensation in the subsequent steps have theoretical support.

[0058] In the optical distortion modeling part, the present invention utilizes the common optical distortion characteristics in the microscope system, including the influence of radial distortion and tangential distortion. Radial distortion is an error caused by the unsatisfactory geometric characteristics in the design and processing of optical elements, which manifests as image geometric distortion, and the degree of distortion increases with the increase of the optical center distance. Its error can be described by the following mathematical model:

[0059] x′=x(1+k1r 2 +k2r 4 +k3r 6 )

[0060] y′=y(1+k1r 2 +k2r 4 +k3r 6 )

[0061] Among them, x, y are the original coordinates not affected by distortion, x′, y′ are the final image coordinates under the effect of optical distortion, is the distance from the image point to the optical center; k1, k2, k3 are radial distortion coefficients, reflecting the nonlinearity of radial distortion. These distortion parameters can be accurately obtained through experimental calibration.

[0062] Tangential distortion is caused by the deviation of the optical axis and the lens assembly error, which manifests as the asymmetry of the image coordinates. The error can be described by the following model:

[0063] x′=x+[2p1xy+p2(r 2 +2x 2 )]

[0064] y′=y+[p1(r 2 +2y 2 )+2p2xy]

[0065] Among them, p1 and p2 are tangential distortion coefficients, corresponding to the distortion effects in different directions. The tangential distortion model can better describe the image coordinate deviation caused by the optical axis offset.

[0066] In order to clearly quantify the total error of optical distortion to the imaging system, the present invention defines the comprehensive error of optical distortion as:

[0067]

[0068] This formula represents the offset of the coordinates of the points before and after optical distortion, and can provide a basis for error quantification for subsequent error decoupling and optimization.

[0069] In the mechanical disturbance modeling part, the present invention establishes a displacement error and tilt error model describing the mechanical disturbance by analyzing the impact of the dynamic environment on the system accuracy. The displacement part of the dynamic mechanical disturbance error can be represented by the following function:

[0070] Δ x (t) = α x cos(ωt)

[0071] Δ y (t) = α y sin(ωt)

[0072] Among them, α x , α y are the amplitudes of mechanical vibration in the x and y directions, respectively, and ω represents the vibration frequency. The above model effectively describes the periodic influence of dynamic mechanical vibration on system accuracy.

[0073] In addition, the error caused by system tilt is manifested as a dynamic offset of the optical axis, and its mathematical model can be expressed as:

[0074] Δx θ =d·tan(θx )

[0075] Δy θ =d·tan(θ y )

[0076] Where d is the effective path distance of the optical axis, θ x ,θ y is the tilt angle of the system in the x and y directions. The tilt error model can be used to analyze the impact of mechanical disturbance on the asymmetry of the optical path.

[0077] The comprehensive error of mechanical disturbance can be expressed by the superposition of displacement error and tilt error as follows:

[0078] Δ x =Δ x (t)+Δx θ

[0079] Δ y =Δ y (t)+Δy θ

[0080] The above formula reflects the time dynamic characteristics of mechanical disturbance and its coupling effect with the system geometric parameters.

[0081] In order to ensure the integrity and accuracy of the modeling of the present invention, this embodiment obtains all the distortion parameters and disturbance parameters involved by experimental calibration method, including optical distortion parameters k1, k2, k3, p1, p2 and mechanical disturbance parameter α x , α y ,ω,θ x ,θ y During the experimental calibration process, the imaging errors of the system under static and dynamic conditions are measured and fitted by using standard calibration targets and high-precision measurement equipment to achieve accurate estimation of the above parameters.

[0082] For step S2, in this embodiment, a joint dynamic error model of optical distortion and mechanical disturbance is constructed. This model is used to describe the coupling effect of optical distortion and mechanical disturbance in the system and its influence on calibration accuracy, and is an important basis for subsequent error decoupling and optimization.

[0083] In actual operation, optical distortion and mechanical disturbance are two main sources of error. Optical distortion is usually a static error, and its characteristics are determined by the design parameters and manufacturing accuracy of the optical components; while mechanical disturbance is a dynamic error, and its characteristics are dynamically affected by factors such as vibration, temperature change, and equipment working status. In order to accurately describe the combined impact of the two on system accuracy, this paper proposes a joint dynamic error model, which integrates optical distortion error and mechanical disturbance error into a unified mathematical framework.

[0084] In order to achieve the coupled description of optical distortion and mechanical disturbance, this embodiment first superimposes the errors of the optical distortion model and the mechanical disturbance model established in the aforementioned step S1 to form a joint error model. The joint dynamic error model describes the total error of the system in a dynamic environment, and its mathematical expression is:

[0085]

[0086] in:

[0087] x′, y′ are the coordinates of the image point after optical distortion;

[0088] x, y are theoretical ideal coordinates;

[0089] Δ x (t) and Δ y (t) is the displacement error component of the mechanical disturbance, caused by external vibration;

[0090] Δx θ and Δy θ is the tilt error component of the mechanical disturbance, caused by the tilt of the system.

[0091] In this model, the optical distortion error is determined by the aforementioned radial distortion and tangential distortion models, which depend on the distortion parameters k1, k2, k3, p1, p2. The mechanical perturbation error component includes the time-dependent dynamic displacement error Δ x (t), Δ y (t) and the geometry-dependent tilt error Δx θ , Δy θ The superposition and coupling between these error components constitute the comprehensive dynamic error of the system.

[0092] The implementation method of the joint dynamic error model includes the following steps:

[0093] First, the distorted image coordinates x', y' in the optical distortion model are obtained. These coordinates are obtained by experimental calibration or calculation with known distortion parameters. The specific calculation formula refers to the definition of the optical distortion model in step S1.

[0094] Secondly, combined with the mechanical disturbance model, the dynamic displacement error component Δ x (t), Δ y (t) and the tilt error component Δx θ , Δy θ The dynamic displacement error component is determined by sensor data or vibration analysis formula, and the tilt error component is determined by the system's geometric parameters (such as optical path distance d) and the dynamic tilt angle Δx θ , Δy θ Calculated.

[0095] Finally, the above error components are substituted into the mathematical expression of the joint dynamic error model to calculate the comprehensive dynamic error value of the system e dynamic (t). This error value is used to quantitatively describe the calibration error of the system under the current dynamic conditions.

[0096] In order to facilitate subsequent processing, the joint dynamic error model can not only describe the size of the error, but also reflect the interactive influence relationship between optical distortion and mechanical disturbance. By separately modeling the influence of different error components, the present invention realizes the unified modeling of optical distortion and mechanical disturbance, and lays a mathematical foundation for subsequent dynamic error decoupling and parameter optimization.

[0097] For step S3, in this embodiment, for the method of improving the optical accuracy of the micro-microscope system and optimizing the image calibration error, the joint dynamic error model is used to decouple and separate the system error, and the dynamic error is separated into a static component and a dynamic component. Through this step, the independent contribution of optical distortion and mechanical disturbance in the error can be clearly identified, providing a basis for subsequent error optimization and real-time compensation.

[0098] Based on the joint dynamic error model, the error decoupling method proposed in the present invention first assumes that the total error consists of a static component caused by optical distortion and a dynamic component caused by mechanical disturbance, that is:

[0099] e dynamic (t) = F static (x,y)+F dynamic (t)

[0100] Among them, F static (x, y) represents the static component caused by optical distortion, which changes with the spatial position of the image coordinates x, y, but has nothing to do with time; F dynamic (t) represents the dynamic component caused by mechanical disturbance, which varies with time but is independent of the spatial coordinates.

[0101] This embodiment realizes error decoupling by using a time-space separation method, and converts the joint dynamic error e dynamic (t) is separated into static and dynamic parts. Specifically, the core of time-space separation is to use the low-frequency part of the error as the static component and the high-frequency part as the dynamic component based on the time and space characteristics of the error signal. The following is a specific description of the implementation process:

[0102] First, the total error signal e is collected in a dynamic environment through experimental measurement or system calibration. dynamic (t). This signal contains the combined errors caused by optical distortion and mechanical disturbance.

[0103] Next, the total error signal is subjected to frequency domain analysis. In this embodiment, the time domain signal is converted to the frequency domain by fast Fourier transform (FFT) to obtain the spectrum distribution of the error signal. The frequency domain conversion formula is:

[0104]

[0105] Where E(f) is the spectrum distribution of the total error, and f is the frequency. Spectral analysis can intuitively display the distribution characteristics of the static component (low-frequency part) and the dynamic component (high-frequency part).

[0106] In order to extract the static component F static (x, y), this embodiment applies a low-pass filter to the spectrum signal E(f), retaining the low-frequency signal and eliminating high-frequency noise. The filter function of the low-pass filter is defined as:

[0107]

[0108] Among them, f cut is the cut-off frequency. Through low-pass filtering, the static component can be expressed as:

[0109]

[0110] Dynamic Component F dynamic (t) is extracted from the total error through a high-pass filter, and the filter function of the high-pass filter is defined as:

[0111]

[0112] The dynamic component is calculated as:

[0113]

[0114] Through the above-mentioned time-space separation process, the static error caused by optical distortion and the dynamic error caused by mechanical disturbance are successfully separated, laying the foundation for the accurate identification and analysis of the error sources.

[0115] In addition, in order to improve the decoupling accuracy, this embodiment also combines the timing filtering technology to separate the dynamic component F dynamic (t) Perform time domain smoothing. The smoothing filter adopts the Kalman filter method, and the specific formula is as follows:

[0116]

[0117] Among them, K t is the Kalman gain, is the predicted value of the dynamic error. Through smoothing filtering, the random noise component in the high-frequency dynamic error is effectively reduced.

[0118] The error decoupling process completed in this step can accurately separate the independent effects of optical distortion and mechanical disturbance, and provide a basis for the error source for parameter optimization and error compensation in subsequent steps. By combining the frequency characteristics and timing characteristics of the error signal, this embodiment significantly improves the accuracy and robustness of error decoupling. This process is of great significance to the dynamic calibration of the system and the optimization of optical accuracy.

[0119] For step S4, the optical accuracy improvement and image calibration error optimization method of the micro-microscope system are optimized based on the decoupled error model, aiming to minimize the system calibration error and provide optimal parameter support for subsequent dynamic compensation.

[0120] In this step, the core of error optimization is to construct an objective function with the minimization of the sum of squared errors as the optimization goal. At the same time, combined with the system constraints, the optical distortion parameters and mechanical disturbance parameters are adjusted to achieve the optimal system calibration accuracy.

[0121] The objective function defined in this embodiment is as follows:

[0122]

[0123] in:

[0124] represents the total error of the i-th measurement point, including optical distortion and mechanical disturbance;

[0125] represents the dynamic error component of the i-th measurement point, which has been obtained by decoupling in step S3;

[0126] N is the total number of measurement points.

[0127] The physical meaning of the objective function J is to minimize the square difference between the dynamic error and the static error, ensuring that both the dynamic error and the static error are effectively optimized.

[0128] In order to ensure the controllability and rationality of the optimization process, this embodiment introduces the following constraints:

[0129] Static error continuity constraint: requires the static error model F static (x, y) has second-order continuous differentiability, that is:

[0130] F static (x,y)∈C 2

[0131] This constraint ensures that the mathematical model of optical distortion has good smoothness and is conducive to the global optimization of system errors.

[0132] Dynamic error amplitude constraint: The amplitude of the dynamic error is required not to exceed the preset threshold, that is:

[0133] ||F dynamic (t)||≤∈

[0134] Among them, ∈ is the upper limit of the amplitude of the dynamic error, and the specific value can be determined according to the dynamic environment and design requirements of the system.

[0135] To achieve the optimal solution of the above objective function, this embodiment uses the gradient optimization method to iteratively update the parameters. The optical distortion parameters involved in the optimization process include radial distortion coefficients k1, k2, k3 and tangential distortion coefficients p1, p2. The mechanical disturbance parameters include the vibration amplitude α x , α y , vibration frequency ω, and tilt angle θ x ,θ y .

[0136] The specific update formula for gradient optimization is:

[0137]

[0138] in:

[0139] θ represents the parameters to be optimized, including optical distortion parameters and mechanical disturbance parameters;

[0140] n is the number of optimization iterations;

[0141] η is the learning rate, which determines the step size of each update;

[0142] is the partial derivative of the objective function with respect to the parameter θ, indicating the influence of the parameter on the error.

[0143] In this embodiment, the calculation of the partial derivative is based on the analytical expression of the error model and is obtained by taking the partial derivative of the objective function J with respect to each parameter.

[0144] The operation process of the optimization algorithm is as follows:

[0145] Initialize the initial values ​​of all parameters θ to be optimized. The initial values ​​can be set based on experimental calibration results.

[0146] In each iteration, the objective function J and partial derivatives are calculated according to the current parameter values

[0147] Update parameter values ​​according to the gradient optimization formula;

[0148] Repeat the above steps until the objective function converges, that is, when the change in the objective function |J (n+1) -J (n) When the value is less than the set convergence threshold, the iteration stops.

[0149] In actual operation, in order to further improve the efficiency and stability of optimization, this embodiment improves the gradient optimization process, including introducing an adaptive learning rate mechanism and a second-order optimization method (such as Newton's method) to accelerate convergence.

[0150] The optimized parameter values ​​can be directly used in the dynamic error compensation model to correct the calibration error in a dynamic environment in real time.

[0151] In this step, the key parameters of the error model are globally optimized by the gradient optimization method, so that the calibration error of the system in a complex dynamic environment is significantly reduced. At the same time, the objective function construction and constraint condition design of this embodiment fully consider the physical characteristics and actual application scenarios of the system, and the optimization result has high accuracy and practicality.

[0152] For step S5, in this embodiment, the optical accuracy improvement and image calibration error optimization method of the micro-microscope system are aimed at obtaining dynamic disturbance data through a real-time sensor, and the image calibration error is dynamically compensated in real time based on the aforementioned optimization results, thereby ensuring that the system can obtain high-precision calibration results and imaging quality in a dynamic environment.

[0153] In order to achieve dynamic compensation, this embodiment first combines the parameters optimized in step S4 to construct a real-time dynamic error compensation model. The model uses real-time sensor data as input, calculates the dynamic error components caused by mechanical disturbances, and compensates for the dynamic calibration error of the system by correcting these error components.

[0154] Real-time sensor data collection is the basis of this embodiment. In order to accurately capture the dynamic errors caused by mechanical disturbances, this embodiment uses a variety of sensors for data collection, including:

[0155] Acceleration sensor, used to measure the vibration acceleration signal of the system in real time, and further calculate the displacement error caused by vibration;

[0156] The gyroscope is used to measure the angular velocity signal of the system in real time and further calculate the error component caused by mechanical tilt.

[0157] The acceleration signal collected by the acceleration sensor is integrated to obtain the displacement error component. The specific formula is as follows:

[0158]

[0159] Among them, a x (τ) and a y (τ) are acceleration signals collected by the sensor in the x and y directions, Δ x (t) and Δ y(t) is the displacement error component in the corresponding direction.

[0160] The angular velocity signal collected by the gyroscope is integrated to obtain the system's tilt angle error, and the tilt angle is further used to calculate the optical axis offset. The specific formula is as follows:

[0161] Δx θ =d·tan(θ x )

[0162] Δy θ =d·tan(θ y )

[0163] Among them, θ x and θ y is the tilt angle collected by the gyroscope, and d is the path length of the system optical axis.

[0164] The real-time dynamic compensation model constructs the comprehensive dynamic error compensation formula of the system by superimposing the displacement error and tilt error components as follows:

[0165] x c = x′-(Δ x (t)+Δx θ )

[0166] y c =y′-(Δ y (t)+Δy θ )

[0167] Among them, x c and c are the calibration coordinates after dynamic compensation, and x′ and y′ are the image point coordinates before compensation.

[0168] The dynamic error compensation operation process of this embodiment is as follows:

[0169] The sensor collects mechanical disturbance data in real time and transmits it to the processing unit at a high sampling rate;

[0170] The processing unit calculates the displacement error component and the tilt error component using the above formula;

[0171] Substitute the error components into the dynamic error compensation formula to calculate the calibrated image coordinates;

[0172] Adjust the imaging parameters of the system in real time to ensure the accuracy of the image coordinates after compensation.

[0173] In order to enhance the accuracy of real-time dynamic compensation, this embodiment combines the time series prediction algorithm to filter and predict the disturbance data. Specifically, the Kalman filter is used to process the sensor signal in real time, and the filtering formula is as follows:

[0174]

[0175] in, is the current predicted value, z(t) is the real-time measured value, K t is the Kalman gain. Kalman filtering can effectively suppress the noise in the sensor signal and improve the accuracy of dynamic compensation.

[0176] The real-time dynamic compensation method of this embodiment can not only effectively correct the dynamic error of the current system, but also has adaptive capabilities, and can adjust the parameters of the compensation model in real time according to changes in the dynamic environment of the system. In addition, combining high-precision sensor data with the above-mentioned optimization results can significantly improve the calibration stability and imaging quality of the system in a dynamic environment.

[0177] Through the real-time dynamic compensation method of this embodiment, the system can quickly respond to the impact of dynamic disturbances and achieve accurate correction of dynamic errors, thereby effectively ensuring the reliability and consistency of system calibration accuracy and imaging effects.

[0178] In general, the present invention establishes an optical distortion model and a mechanical disturbance model, constructs a joint dynamic error model, decouples and separates errors, and iteratively optimizes optical distortion parameters and mechanical disturbance parameters based on an optimization algorithm, and finally realizes dynamic error compensation in combination with real-time sensor data. This method can effectively separate and correct optical and mechanical errors in complex dynamic environments, significantly improve the system calibration accuracy, ensure high accuracy and stability of imaging results, and has broad application prospects, especially suitable for dynamic calibration and precision measurement scenarios of microscopic imaging equipment.

[0179] The system for improving the optical accuracy of a micro-microscope system and optimizing the image calibration error described below and the method for improving the optical accuracy of a micro-microscope system and optimizing the image calibration error described above can be referred to each other.

[0180] Please see attached Figure 2 The present invention also provides a system for improving the optical accuracy of a micro-microscope system and optimizing image calibration errors, comprising:

[0181] An optical distortion modeling module 10 is used to establish a model describing radial distortion and tangential distortion;

[0182] A mechanical disturbance modeling module 20, for establishing a displacement and tilt model describing dynamic disturbances;

[0183] An error decoupling module 30, for performing spatiotemporal decomposition and frequency domain analysis on the joint dynamic error;

[0184] An optimization module 40, used for optimizing distortion parameters and disturbance parameters based on an objective function and constraint conditions;

[0185] A real-time compensation module 50 is used to compensate and calibrate the dynamic disturbance according to the real-time sensor data;

[0186] The data acquisition module 60 is used to obtain the dynamic disturbance signal and the real-time data of the optical system.

[0187] The device of this embodiment can be used to execute the above method embodiment, and its principles and technical effects are similar, which will not be repeated here.

[0188] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for improving the optical accuracy of a micro-microscope system and optimizing image calibration errors, characterized in that: The following steps are involved: Establish an optical distortion model and a mechanical disturbance model to describe the static error caused by optical distortion and the dynamic error caused by mechanical disturbance respectively; Construct a joint dynamic error model of optical distortion and mechanical disturbance; Decouple and separate the joint dynamic error, and separate the dynamic error into static and dynamic components; Based on the decoupled error model, the optical distortion parameters and mechanical disturbance parameters are optimized to minimize the system calibration error. Dynamic disturbance data is acquired through real-time sensors, and image calibration errors are compensated in real time based on the optimization results to obtain calibrated high-precision images.

2. The method for improving the optical accuracy of a micro-microscope system and optimizing image calibration errors according to claim 1, characterized in that: The optical distortion model includes radial distortion and tangential distortion. The optical distortion error is determined by multiple distortion parameters, including at least radial distortion parameters and tangential distortion parameters. The radial distortion error increases nonlinearly with increasing optical center distance.

3. The method for improving the optical accuracy of a micro-microscope system and optimizing image calibration errors according to claim 1, characterized in that: The mechanical disturbance model is used to describe the displacement and tilt errors of the system in a dynamic environment, including: Dynamic displacement error, which is caused by external vibration and manifests itself as amplitude changes along the optical axis and dynamic fluctuations of vibration frequency; Tilt angle error, which is caused by the tilt of mechanical components, manifests itself as an angular deviation of the optical axis of the optical system during the imaging process, resulting in a deviation in the optical path of the imaging area.

4. The method for improving the optical accuracy of a micro-microscope system and optimizing image calibration errors according to claim 1, characterized in that: The step of decoupling and separating the joint dynamic error comprises: Decompose the dynamic error into static error and dynamic error by using time-space decomposition; Frequency domain analysis is used to extract dynamic error components through high-pass filtering and static error components through low-pass filtering.

5. The method for improving the optical accuracy of a micro-microscope system and optimizing image calibration errors according to claim 4, characterized in that: In the decoupling process of the dynamic error, the multi-dimensional sensor data of the mechanical disturbance signal is fused and analyzed to improve the accuracy of error separation, wherein the multi-dimensional sensor data includes the time series information of acceleration, angular velocity and displacement, and the fusion method adopts weighted filtering or Kalman filtering method.

6. The method for improving the optical accuracy of a micro-microscope system and optimizing image calibration errors according to claim 1, characterized in that: The step of optimizing the optical distortion parameters and the mechanical disturbance parameters and minimizing the system calibration error is implemented based on minimizing the objective function, where the objective function is the minimum value of the sum of the squares of the dynamic error and the static error, and the optimization constraints include: Static error continuity constraint requires the static error model to be second-order continuously differentiable; The dynamic error amplitude constraint requires that the dynamic error component does not exceed a preset threshold.

7. The method for improving the optical accuracy of a micro-microscope system and optimizing image calibration errors according to claim 1, characterized in that: The error optimization process is solved by the gradient descent method, and includes the following steps: Initialize optical distortion parameters and mechanical perturbation parameters; Calculate the total error of the current system; Calculate the gradient of each parameter according to the error function; The parameters are iteratively updated based on the gradient until a preset convergence condition is reached.

8. The method for improving the optical accuracy of a micro-microscope system and optimizing image calibration errors according to claim 1, characterized in that: The steps of real-time dynamic compensation of image calibration error are based on dynamic disturbance data acquired by the sensor, and include: Acquire the displacement and tilt data of mechanical disturbance in real time through acceleration sensor and gyroscope; Dynamically calibrate and compensate image coordinates according to mechanical disturbance data; The compensated image coordinates are input into the system for high-precision image reconstruction.

9. The method for improving the optical accuracy of a micro-microscope system and optimizing image calibration errors according to claim 1, characterized in that: The compensation model of the real-time dynamic compensation dynamically adjusts the compensation coefficient according to the frequency and amplitude of the dynamic disturbance to eliminate the influence of high-frequency vibration on the image.

10. A system for improving the optical accuracy of a micro-microscope system and optimizing image calibration errors, used to perform the method according to any one of claims 1 to 9, characterized in that: include: Optical distortion modeling module, used to build models describing radial distortion and tangential distortion; Mechanical disturbance modeling module, used to build displacement and tilt models to describe dynamic disturbances; Error decoupling module, used for time-space decomposition and frequency-domain analysis of joint dynamic errors; An optimization module, used to optimize distortion parameters and perturbation parameters based on objective functions and constraints; A real-time compensation module is used to compensate and calibrate dynamic disturbances based on real-time sensor data; The data acquisition module is used to obtain dynamic disturbance signals and real-time data of the optical system.

Images