An optical stability evaluation and prediction method based on an improved LSTM model

Abstract

The application relates to the technical field of optical microscope imaging, and discloses an optical stability evaluation and prediction method based on an improved LSTM model, which realizes accurate evaluation of imaging stability by acquiring imaging data, environmental sensor data and running state data of the optical microscope to be measured and performing multi-source data fusion analysis by using an improved LSTM model. The improved LSTM model can accurately identify the root cause of unstable imaging, significantly improves the accuracy and reliability of the evaluation result, adopts a dynamic weight adjustment strategy and real-time data processing technology, enables the model to quickly respond to environmental changes and equipment state fluctuations, realizes efficient prediction of the imaging stability, solves the problem of poor real-time performance of the traditional method, combines the real-time evaluation and prediction results, further provides an intelligent early warning function, helps researchers to find and solve unstable problems in the imaging process in advance, effectively prevents the imaging quality from being reduced, and reduces experimental uncertainty.

Description

Technical Field

[0001] This invention relates to the field of optical microscope imaging technology, and in particular to an optical stability assessment and prediction method based on an improved LSTM model. Background Technology

[0002] In scientific research applications of optical microscopes, especially high-resolution electron microscopes, imaging stability is directly related to the accuracy and reliability of observation results. However, various factors such as environmental factors (e.g., changes in temperature, humidity, and magnetic fields), equipment vibration, and power supply fluctuations can adversely affect imaging stability, leading to a decrease in image quality.

[0003] Traditional methods primarily rely on manual monitoring and post-processing adjustments to maintain imaging stability. Manual monitoring involves researchers periodically checking environmental parameters (such as temperature and humidity) and equipment status, while post-processing involves adjusting microscope parameters or environmental conditions based on experience after image quality degradation is detected. However, these methods have significant drawbacks: First, they lack real-time capability, failing to achieve continuous monitoring and often only being discovered after problems occur; second, they have low accuracy, relying on subjective judgment and making it difficult to accurately identify the root cause of problems; furthermore, traditional methods lack predictive capabilities, failing to prevent imaging instability issues in advance, leading to increased experimental uncertainty. Simultaneously, manual monitoring and adjustments require substantial manpower and have limited ability to cope with complex and changing environmental conditions, hindering rapid response and adaptation. These limitations render traditional methods increasingly inadequate for high-precision scientific research applications, necessitating more advanced automated solutions to improve imaging stability and experimental efficiency. Summary of the Invention

[0004] This invention provides an optical stability assessment and prediction method based on an improved LSTM model, which addresses the shortcomings of existing optical stability assessment and prediction methods based on improved LSTM models, such as poor real-time performance, low accuracy, and lack of preventative capabilities.

[0005] This invention provides an optical stability assessment and prediction method based on an improved LSTM model, comprising:

[0006] Acquire imaging data, environmental sensor data, and operational status data from the optical microscope under test;

[0007] Based on the imaging data, environmental sensor data, and operational status data of the optical microscope under test, the imaging stability evaluation results of the optical microscope under test are obtained through the optical microscope imaging stability evaluation model, which is an improved LSTM model.

[0008] According to the optical stability assessment and prediction method based on an improved LSTM model provided by the present invention, the environmental sensor data includes any one or any combination of the following: temperature, humidity, vibration, magnetic field, and noise.

[0009] According to the present invention, an optical stability assessment and prediction method based on an improved LSTM model is provided, and the operating status data includes light source current, vacuum value, and liquid nitrogen content.

[0010] According to the present invention, an optical stability assessment and prediction method based on an improved LSTM model is provided. The method involves obtaining the imaging stability assessment result of the optical microscope under test based on the imaging data, environmental sensor data, and operational status data of the microscope under test, using an optical microscope imaging stability assessment model. The method includes:

[0011] Environmental sensor data and operational status data are preprocessed, including any one or any combination of the following: data cleaning, feature extraction, and data normalization.

[0012] According to the optical stability assessment and prediction method based on an improved LSTM model provided by the present invention, when the preprocessing is feature extraction, the step of obtaining the imaging stability assessment result of the optical microscope under test based on the imaging data, environmental sensor data, and operating status data of the optical microscope under test through the optical microscope imaging stability assessment model includes:

[0013] Based on the imaging data, environmental sensor data, and operating status data of the optical microscope under test, characteristic data that affect imaging stability are obtained. Among them, the characteristic data that affect imaging stability include any one of the following or any combination thereof: temperature fluctuation range, temperature change trend, humidity fluctuation range, humidity change trend, peak vibration acceleration, vibration frequency, light source current stability, and image sharpness score.

[0014] According to the present invention, an optical stability assessment and prediction method based on an improved LSTM model is provided, wherein characteristic data affecting imaging stability are obtained based on imaging data of the optical microscope under test, environmental sensor data, and operating status data, including:

[0015] Based on the environmental sensor data of the optical microscope under test, the difference between the maximum and minimum temperature values ​​over the past n time steps is calculated to obtain the temperature fluctuation range of the optical microscope under test over the past n time steps.

[0016] Based on the environmental sensor data of the optical microscope under test, the temperature difference between adjacent time steps is calculated to obtain the temperature change trend.

[0017] According to the present invention, an optical stability assessment and prediction method based on an improved LSTM model is provided, wherein characteristic data affecting imaging stability are obtained based on imaging data of the optical microscope under test, environmental sensor data, and operating status data, including:

[0018] Based on the environmental sensor data of the optical microscope under test, the difference between the maximum and minimum humidity values ​​over the past n time steps is calculated to obtain the humidity fluctuation range of the optical microscope under test over the past n time steps.

[0019] Based on the environmental sensor data of the optical microscope under test, the humidity difference between adjacent time steps is calculated to obtain the humidity change trend.

[0020] According to the present invention, an optical stability assessment and prediction method based on an improved LSTM model is provided, wherein characteristic data affecting imaging stability are obtained based on imaging data of the optical microscope under test, environmental sensor data, and operating status data, including:

[0021] Based on the environmental sensor data of the optical microscope under test, the maximum vibration acceleration of the optical microscope under test at n time steps is obtained, which is taken as the peak vibration acceleration of the optical microscope under test at n time steps.

[0022] According to the present invention, an optical stability assessment and prediction method based on an improved LSTM model is provided, wherein characteristic data affecting imaging stability are obtained based on imaging data of the optical microscope under test, environmental sensor data, and operating status data, including:

[0023] Based on the imaging data of the optical microscope under test, environmental sensor data, and operating status data, the time-domain vibration signal is converted into a frequency-domain signal through fast Fourier transform to obtain the main vibration frequency components, which are used as vibration frequency characteristics.

[0024] According to the present invention, an optical stability assessment and prediction method based on an improved LSTM model is provided, wherein characteristic data affecting imaging stability are obtained based on imaging data of the optical microscope under test, environmental sensor data, and operating status data, including:

[0025] Based on the imaging data of the optical microscope under test, the Laplacian operator is used to perform edge detection on the imaging data to obtain the image edge intensity of the optical microscope under test, which is used as the image sharpness score.

[0026] According to the present invention, an optical stability assessment and prediction method based on an improved LSTM model is provided. The optical microscope imaging stability assessment model includes a stacked LSTM layer, a feature stitching module, and a fully connected layer containing a Dropout mechanism. An attention mechanism layer is provided between adjacent LSTM layers in the stacked LSTM layer. The fully connected layer includes two hidden layers and an output layer. The Dropout mechanism applies to the two hidden layers. The first LSTM layer in the stacked LSTM layer is configured to receive environmental sensor data and operating status data of the optical microscope under test. The output of the first LSTM layer in the stacked LSTM layer serves as the input to the feature stitching module. The output of the feature stitching module serves as the input to the hidden layer of the fully connected layer. The output of the hidden layer in the fully connected layer serves as the input to its output layer. The output layer of the fully connected layer is configured to output the imaging stability assessment result of the optical microscope under test.

[0027] According to the present invention, an optical stability assessment and prediction method based on an improved LSTM model is provided. The method involves obtaining the imaging stability assessment result of the optical microscope under test based on the imaging data, environmental sensor data, and operational status data of the microscope under test, using an optical microscope imaging stability assessment model. The method includes:

[0028] Based on the imaging data, environmental sensor data, and operational status data of the optical microscope under test, the imaging stability evaluation score of the optical microscope under test is obtained by using the imaging stability evaluation expression through the optical microscope imaging stability evaluation model.

[0029] The imaging stability evaluation score of the optical microscope under test is compared with the preset range to obtain the imaging stability evaluation result of the optical microscope under test.

[0030] According to the optical stability assessment and prediction method based on an improved LSTM model provided by the present invention, the imaging stability assessment expression is as follows:

[0031]

[0032] In the imaging stability evaluation expression, A1, B2, ..., Z N The data represents the feature data that affects the imaging stability of the optical microscope under test, including imaging data, environmental sensor data, and operational status data. α, β, ..., ζ represent the dynamic weights of the corresponding feature data, and S represents the imaging stability evaluation score of the optical microscope under test.

[0033] According to the present invention, an optical stability assessment and prediction method based on an improved LSTM model is provided, wherein comparing the imaging stability assessment score of the optical microscope under test with a preset range to obtain the imaging stability assessment result of the optical microscope under test includes:

[0034] When the imaging stability evaluation score of the optical microscope under test is outside the preset range and less than the minimum value within the preset range, the imaging stability evaluation result of the optical microscope under test is determined to be a decrease in imaging stability. Based on the environmental sensor data and operating status data of the optical microscope under test, the influencing factors affecting imaging stability are obtained, and an early warning signal and the influencing factors affecting imaging stability are issued.

[0035] When the imaging stability evaluation score of the optical microscope under test is within the preset range, the imaging stability evaluation result of the optical microscope under test is judged to be passing.

[0036] When the imaging stability evaluation score of the optical microscope under test is outside the preset range but higher than the maximum value within the preset range, the imaging stability evaluation result of the optical microscope under test is determined to be an increase in imaging stability.

[0037] This invention also provides an optical stability assessment and prediction system based on an improved LSTM model, comprising:

[0038] The data acquisition module is used to acquire imaging data, environmental sensor data, and operating status data of the optical microscope under test.

[0039] The optical microscope imaging stability assessment module is used to: obtain the imaging stability assessment results of the optical microscope under test based on the imaging data, environmental sensor data and operating status data of the optical microscope under test, through the optical microscope imaging stability assessment model, wherein the optical microscope imaging stability assessment model is an improved LSTM model.

[0040] The present invention also provides an electronic device, including a processor and a memory storing a computer program, wherein the processor executes the computer program to implement any of the above-described optical stability assessment and prediction methods based on an improved LSTM model.

[0041] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described optical stability assessment and prediction methods based on an improved LSTM model.

[0042] The present invention also provides a computer program product, the computer program product including a computer program that can be stored on a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer is able to execute any of the above-described optical stability assessment and prediction methods based on the improved LSTM model.

[0043] This invention provides an optical stability assessment and prediction method based on an improved LSTM model. By acquiring imaging data, environmental sensor data, and operational status data from the optical microscope under test, and utilizing an improved LSTM model for multi-source data fusion analysis, it achieves accurate assessment of imaging stability. Compared to traditional manual monitoring and post-processing adjustments, this invention offers significant advantages: First, the application of the improved LSTM model accurately identifies the root causes of imaging instability, significantly improving the accuracy and reliability of the assessment results. Second, the use of a dynamic weight adjustment strategy and real-time data processing technology enables the model to quickly respond to environmental changes and equipment status fluctuations, achieving efficient prediction of imaging stability and solving the problem of poor real-time performance in traditional methods. Furthermore, combining real-time assessment and prediction results, this invention also provides an intelligent early warning function, helping researchers to identify and resolve instability issues during the imaging process in advance, thereby effectively preventing image quality degradation and reducing experimental uncertainty. Simultaneously, this invention significantly reduces the need for manual monitoring and adjustments through automation technology, lowering labor costs and demonstrating stronger adaptability in complex and changing environmental conditions. These advantages enable this invention to significantly improve imaging stability and experimental efficiency in high-precision scientific applications, overcoming the limitations of traditional methods. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0045] Figure 1 This is a flowchart illustrating an optical stability assessment and prediction method based on an improved LSTM model provided by the present invention.

[0046] Figure 2 This is a schematic diagram of the architecture of an optical microscope imaging stability evaluation model.

[0047] Figure 3 This is a schematic diagram of the structure of an optical stability assessment and prediction system based on an improved LSTM model provided by the present invention.

[0048] Figure 4 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, embodiments of this invention, and should not be construed as limiting the invention. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention. In the description of this invention, it should be understood that the terminology used is for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0050] Figure 1 This is a flowchart illustrating an optical stability assessment and prediction method based on an improved LSTM model provided by the present invention. The execution entity of this optical stability assessment and prediction method based on an improved LSTM model can be any applicable terminal-side device or network-side device, such as an optical microscope imaging stability assessment device.

[0051] See Figure 1 The present invention provides an optical stability assessment and prediction method based on an improved LSTM model, which may include:

[0052] S110. Acquire imaging data, environmental sensor data, and operating status data of the optical microscope under test. The imaging data is the image obtained after imaging with the optical microscope under test. In this embodiment, the environmental sensor data includes temperature, humidity, and vibration, and the operating status data includes the light source current.

[0053] In one embodiment, after obtaining the imaging data, environmental sensor data, and operating status data of the optical microscope under test, S110 can preprocess the data. The preprocessing includes any one or any combination of the following: data cleaning, feature extraction, and data normalization.

[0054] In one embodiment, data cleaning may include removing noise, outliers, and duplicate records from the imaging data, environmental sensor data, and operational status data of the optical microscope under test, to ensure the accuracy and consistency of the data.

[0055] In one embodiment, feature extraction may be based on imaging data of the optical microscope under test, environmental sensor data, and operating status data to obtain feature data that affects imaging stability (e.g., directly or indirectly causing image quality fluctuations (such as resolution reduction, signal-to-noise ratio reduction, or drift) during the optical microscope imaging process), so as to provide an important basis for model prediction. In this embodiment, the feature data that affects imaging stability includes: temperature fluctuation range, temperature change trend, humidity fluctuation range, humidity change trend, peak vibration acceleration, vibration frequency, and image sharpness score.

[0056] In this embodiment, based on the environmental sensor data of the optical microscope under test, the difference between the maximum and minimum temperatures over the past n time steps is calculated to obtain the temperature fluctuation range of the optical microscope under test over the past n time steps. The expression can be: T_range=max(T_1,T_2,...,T_n)-min(T_1,T_2,...,T_n), where T_i is the temperature value at the i-th time step; based on the environmental sensor data of the optical microscope under test, the temperature difference between adjacent time steps is calculated to obtain the temperature change trend. The expression can be: T_rate=T_i-T_{i-1}.

[0057] In this embodiment, similarly, based on the environmental sensor data of the optical microscope under test, the difference between the maximum and minimum humidity values ​​within the past n time steps is calculated to obtain the humidity fluctuation range of the optical microscope under test over the past n time steps; based on the environmental sensor data of the optical microscope under test, the humidity difference between adjacent time steps is calculated to obtain the humidity change trend.

[0058] In this embodiment, based on the environmental sensor data of the optical microscope under test, the maximum vibration acceleration of the optical microscope under test at n time steps is obtained, which is taken as the peak vibration acceleration of the optical microscope under test at n time steps. The expression can be: A_peak=max(A_1,A_2,...,A_n), where A_i is the vibration acceleration value at the i-th time step.

[0059] In this embodiment, based on the imaging data, environmental sensor data, and operating status data of the optical microscope under test, the time-domain vibration signal is converted into a frequency-domain signal by Fast Fourier Transform (FFT) to obtain the main vibration frequency components, which are used as vibration frequency characteristics.

[0060] In this embodiment, based on the imaging data of the optical microscope under test, edge detection is performed on the imaging data using the Laplacian operator to obtain the image edge intensity of the optical microscope under test, which is used as the image sharpness score. The expression can be: in This represents the gradient value of the image at pixel (x, y). A larger gradient indicates a sharper image, and vice versa. By extracting image sharpness features, image quality can be quantitatively evaluated.

[0061] In one embodiment, data normalization may involve converting feature values ​​of different dimensions and ranges into a uniform scale for model processing. This embodiment employs a min-max normalization method to normalize the feature data, converting feature values ​​of different dimensions and ranges to the [0, 1] interval. The formula is x' = (x - x_min) / (x_max - x_min), where x is the original feature value, x_min and x_max are the minimum and maximum values ​​of the feature data, respectively, and x' is the normalized feature value.

[0062] S120. Based on the imaging data, environmental sensor data, and operating status data of the optical microscope under test, the imaging stability evaluation result of the optical microscope under test is obtained through the optical microscope imaging stability evaluation model. The optical microscope imaging stability evaluation model is an improved LSTM model.

[0063] In one embodiment, the optical microscope imaging stability evaluation model includes a stacked LSTM layer, a feature stitching module, and a fully connected layer containing a Dropout mechanism. An attention mechanism layer is provided between adjacent LSTM layers in the stacked LSTM layer. The fully connected layer includes two hidden layers and an output layer. The Dropout mechanism applies to the two hidden layers. The first LSTM layer in the stacked LSTM layer is configured to receive environmental sensor data and operating status data of the optical microscope under test. The output of the first LSTM layer in the stacked LSTM layer serves as the input to the feature stitching module. The output of the feature stitching module serves as the input to the hidden layer of the fully connected layer. The output of the hidden layer in the fully connected layer serves as the input to its output layer. The output layer of the fully connected layer is configured to output the imaging stability evaluation result of the optical microscope under test.

[0064] This embodiment uses stacked LSTM layers to build a deeper network structure for the optical microscope imaging stability evaluation model, thereby improving the model's learning ability and nonlinear expression ability; it introduces an attention mechanism to enhance the model's ability to capture key information; and it uses techniques such as Dropout to prevent overfitting and improve the model's generalization ability.

[0065] In this embodiment, the stacked LSTM layers employed in this invention are specifically a three-layer LSTM structure. The first LSTM layer contains 50 neurons, used for initial extraction of short-term features from the time-series data; the second LSTM layer contains 30 neurons, further mining mid-term features from the data; and the third LSTM layer contains 20 neurons, focusing on extracting long-term features. This hierarchical feature extraction method can more comprehensively capture the time-series information in the input data (imaging data from the optical microscope under test, environmental sensor data, and operational status data). When constructing the stacked LSTM layers, a layer-by-layer training method is adopted. First, the first LSTM layer is trained to learn the basic patterns of the data. Then, the output of the first LSTM layer is used as the input to the second LSTM layer for training, and so on. Finally, all three LSTM layers are fine-tuned together to improve the overall performance of the model.

[0066] In the optical microscope imaging stability evaluation model, the input is a time series X composed of tx multidimensional vectors x, where tx is the input time step size, denoted as:

[0067] X = (x1, x2, x3, ..., x...) tx )

[0068] The output consists of ty time series prediction results composed of numbers from 0 to 1. ty is the output timing step size, denoted as:

[0069]

[0070] In one embodiment, considering that the influence of various factors on imaging stability may change over different time periods, this embodiment proposes a dynamic weight adjustment strategy. This strategy dynamically adjusts the weights of each input feature based on the changing trends of historical data and the current environmental state. For example, when the temperature value monitored in real time deviates from its normal changing trend and causes a decrease in imaging quality, the weight of the temperature feature is increased to make the model more adaptable to the actual situation and improve the accuracy of prediction.

[0071] In one embodiment, S120 may include:

[0072] Based on the imaging data, environmental sensor data, and operational status data of the optical microscope under test, an imaging stability evaluation score is obtained using the optical microscope imaging stability evaluation model and the imaging stability evaluation expression. The imaging stability evaluation expression is as follows:

[0073]

[0074] In the imaging stability evaluation expression, A1, B2, ..., Z NThe data represents the feature data that affects the imaging stability of the optical microscope under test, including imaging data, environmental sensor data, and operating status data. α,β,….,ζ represent the dynamic weights of the corresponding feature data, and S represents the imaging stability evaluation score of the optical microscope under test.

[0075] The imaging stability evaluation score of the optical microscope under test is compared with a preset range (e.g., The imaging stability evaluation result of the optical microscope under test is obtained by comparison. In this embodiment, when the imaging stability evaluation score of the optical microscope under test is outside the preset range and less than the minimum value within the preset range, the imaging stability evaluation result of the optical microscope under test is determined to be a decrease in imaging stability. Based on the environmental sensor data and operating status data of the optical microscope under test, the influencing factors affecting imaging stability are identified, and a warning signal and the influencing factors affecting imaging stability are issued. When the imaging stability evaluation score of the optical microscope under test is within the preset range, the imaging stability evaluation result of the optical microscope under test is determined to be passing. When the imaging stability evaluation score of the optical microscope under test is outside the preset range and higher than the maximum value within the preset range, the imaging stability evaluation result of the optical microscope under test is determined to be an increase in imaging stability. This embodiment can also predict future scores based on the real-time scores obtained from real-time evaluation and the real-time scores of the past n time steps.

[0076] This invention provides an optical stability assessment and prediction method based on an improved LSTM model. By acquiring imaging data, environmental sensor data, and operational status data from the optical microscope under test, and utilizing an improved LSTM model for multi-source data fusion analysis, it achieves accurate assessment of imaging stability. Compared to traditional manual monitoring and post-processing adjustments, this invention offers significant advantages: First, the application of the improved LSTM model accurately identifies the root causes of imaging instability, significantly improving the accuracy and reliability of the assessment results. Second, the use of a dynamic weight adjustment strategy and real-time data processing technology enables the model to quickly respond to environmental changes and equipment status fluctuations, achieving efficient prediction of imaging stability and solving the problem of poor real-time performance in traditional methods. Furthermore, combining real-time assessment and prediction results, this invention also provides an intelligent early warning function, helping researchers to identify and resolve instability issues during the imaging process in advance, thereby effectively preventing image quality degradation and reducing experimental uncertainty. Simultaneously, this invention significantly reduces the need for manual monitoring and adjustments through automation technology, lowering labor costs and demonstrating stronger adaptability in complex and changing environmental conditions. These advantages enable this invention to significantly improve imaging stability and experimental efficiency in high-precision scientific applications, overcoming the limitations of traditional methods.

[0077] The optical stability assessment and prediction system based on the improved LSTM model provided by the present invention will be described below. The optical stability assessment and prediction system based on the improved LSTM model described below can be referred to in correspondence with the optical stability assessment and prediction method based on the improved LSTM model described above.

[0078] See Figure 3 The present invention provides an optical stability assessment and prediction system based on an improved LSTM model, which may include:

[0079] The data acquisition module is used to acquire imaging data, environmental sensor data, and operating status data of the optical microscope under test.

[0080] The optical microscope imaging stability assessment module is used to: obtain the imaging stability assessment results of the optical microscope under test based on the imaging data, environmental sensor data and operating status data of the optical microscope under test, through the optical microscope imaging stability assessment model, wherein the optical microscope imaging stability assessment model is an improved LSTM model.

[0081] By deploying an optical microscope imaging stability assessment model into the system, an optical stability assessment and prediction system based on an improved LSTM model is formed. This system can receive imaging data, environmental sensor data, and operational status data from the optical microscope under test in real time, and perform online assessment and prediction. When a decrease in imaging stability is predicted, the system immediately issues an early warning signal and provides specific influencing factors (such as excessive ambient noise), thereby helping researchers take timely measures to ensure stable and reliable imaging quality.

[0082] Figure 4 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 4 As shown, the electronic device may include a processor 810, a communications interface 820, a memory 830, and a communication bus 840, wherein the processor 810, the communications interface 820, and the memory 830 communicate with each other via the communication bus 840. The processor 810 can call logical instructions in the memory 830 to execute the following steps:

[0083] Acquire imaging data, environmental sensor data, and operational status data from the optical microscope under test;

[0084] Based on the imaging data, environmental sensor data, and operational status data of the optical microscope under test, the imaging stability evaluation results of the optical microscope under test are obtained through the optical microscope imaging stability evaluation model, which is an improved LSTM model.

[0085] Furthermore, the logical instructions in the aforementioned memory 830 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0086] On the other hand, the present invention also provides a computer program product, the computer program product comprising a computer program, the computer program being able to be stored on a non-transitory computer-readable storage medium, and the computer program being executed by a processor, enabling the computer to perform the following steps:

[0087] Acquire imaging data, environmental sensor data, and operational status data from the optical microscope under test;

[0088] Based on the imaging data, environmental sensor data, and operational status data of the optical microscope under test, the imaging stability evaluation results of the optical microscope under test are obtained through the optical microscope imaging stability evaluation model, which is an improved LSTM model.

[0089] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0090] Acquire imaging data, environmental sensor data, and operational status data from the optical microscope under test;

[0091] Based on the imaging data, environmental sensor data, and operational status data of the optical microscope under test, the imaging stability evaluation results of the optical microscope under test are obtained through the optical microscope imaging stability evaluation model, which is an improved LSTM model.

[0092] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0093] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0094] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for evaluating and predicting optical stability based on an improved LSTM model, characterized in that, include: Acquire imaging data, environmental sensor data, and operational status data from the optical microscope under test; Based on the imaging data, environmental sensor data, and operational status data of the optical microscope under test, the imaging stability evaluation results of the optical microscope under test are obtained through the optical microscope imaging stability evaluation model. The optical microscope imaging stability evaluation model is an improved LSTM model. Environmental sensor data includes any one or any combination of the following: temperature, humidity, vibration, magnetic field, and noise; operating status data includes light source current, vacuum value, and liquid nitrogen content. The process involves using the imaging data, environmental sensor data, and operational status data of the optical microscope under test, and employing an optical microscope imaging stability evaluation model to obtain the imaging stability evaluation results of the optical microscope under test, including: Preprocessing of environmental sensor data and operational status data includes any one or any combination of the following: data cleaning, feature extraction, and data normalization. When the preprocessing is feature extraction, the imaging stability evaluation result of the optical microscope under test is obtained based on the imaging data, environmental sensor data, and operating status data of the optical microscope under test, through the optical microscope imaging stability evaluation model, including: Based on the imaging data, environmental sensor data, and operating status data of the optical microscope under test, characteristic data that affect imaging stability are obtained. Among them, the characteristic data that affect imaging stability include any one of the following or any combination thereof: temperature fluctuation range, temperature change trend, humidity fluctuation range, humidity change trend, peak vibration acceleration, vibration frequency, light source current stability, and image sharpness score. The process involves obtaining characteristic data that affects imaging stability based on the imaging data, environmental sensor data, and operational status data of the optical microscope under test, including: Based on the imaging data of the optical microscope under test, the Laplacian operator is used to perform edge detection on the imaging data to obtain the image edge intensity of the optical microscope under test, which is used as the image sharpness score.

2. The optical stability assessment and prediction method based on the improved LSTM model according to claim 1, characterized in that, The process involves obtaining characteristic data that affects imaging stability based on the imaging data, environmental sensor data, and operational status data of the optical microscope under test, including: Based on the environmental sensor data of the optical microscope under test, the difference between the maximum and minimum temperature values ​​over the past n time steps is calculated to obtain the temperature fluctuation range of the optical microscope under test over the past n time steps. Based on the environmental sensor data of the optical microscope under test, the temperature difference between adjacent time steps is calculated to obtain the temperature change trend.

3. The optical stability assessment and prediction method based on the improved LSTM model according to claim 1, characterized in that, The process involves obtaining characteristic data that affects imaging stability based on the imaging data, environmental sensor data, and operational status data of the optical microscope under test, including: Based on the environmental sensor data of the optical microscope under test, the difference between the maximum and minimum humidity values ​​over the past n time steps is calculated to obtain the humidity fluctuation range of the optical microscope under test over the past n time steps. Based on the environmental sensor data of the optical microscope under test, the humidity difference between adjacent time steps is calculated to obtain the humidity change trend.

4. The optical stability assessment and prediction method based on the improved LSTM model according to claim 1, characterized in that, The process involves obtaining characteristic data that affects imaging stability based on the imaging data, environmental sensor data, and operational status data of the optical microscope under test, including: Based on the environmental sensor data of the optical microscope under test, the maximum vibration acceleration of the optical microscope under test at n time steps is obtained, which is taken as the peak vibration acceleration of the optical microscope under test at n time steps.

5. The optical stability assessment and prediction method based on the improved LSTM model according to claim 1, characterized in that, The process involves obtaining characteristic data that affects imaging stability based on the imaging data, environmental sensor data, and operational status data of the optical microscope under test, including: Based on the imaging data of the optical microscope under test, environmental sensor data, and operating status data, the time-domain vibration signal is converted into a frequency-domain signal through fast Fourier transform to obtain the main vibration frequency components, which are used as vibration frequency characteristics.

6. The optical stability assessment and prediction method based on an improved LSTM model according to any one of claims 1-5, characterized in that, The optical microscope imaging stability evaluation model includes a stacked LSTM layer, a feature stitching module, and a fully connected layer with a Dropout mechanism. An attention mechanism layer is placed between adjacent LSTM layers in the stacked LSTM layer. The fully connected layer includes two hidden layers and an output layer. The Dropout mechanism applies to the two hidden layers. The first LSTM layer in the stacked LSTM layer is configured to receive environmental sensor data and operational status data from the optical microscope under test. The output of the first LSTM layer in the stacked LSTM layer serves as the input to the feature stitching module. The output of the feature stitching module serves as the input to the hidden layer of the fully connected layer. The output of the hidden layer in the fully connected layer serves as the input to its output layer. The output layer of the fully connected layer is configured to output the imaging stability evaluation result of the optical microscope under test.

7. The optical stability assessment and prediction method based on an improved LSTM model according to any one of claims 1-5, characterized in that, The process involves using the imaging data, environmental sensor data, and operational status data of the optical microscope under test, and employing an optical microscope imaging stability evaluation model to obtain the imaging stability evaluation results of the optical microscope under test, including: Based on the imaging data, environmental sensor data, and operational status data of the optical microscope under test, the imaging stability evaluation score of the optical microscope under test is obtained by using the imaging stability evaluation expression through the optical microscope imaging stability evaluation model. The imaging stability evaluation score of the optical microscope under test is compared with the preset range to obtain the imaging stability evaluation result of the optical microscope under test.

8. The optical stability assessment and prediction method based on the improved LSTM model according to claim 7, characterized in that, The expression for evaluating imaging stability is: In the imaging stability evaluation expression, This refers to the characteristic data that affects the imaging stability of the optical microscope under test, including imaging data, environmental sensor data, and operational status data. The dynamic weights of the corresponding feature data are represented by , and S represents the imaging stability evaluation score of the optical microscope under test.

9. The optical stability assessment and prediction method based on the improved LSTM model according to claim 7, characterized in that, The step of comparing the imaging stability evaluation score of the optical microscope under test with a preset range to obtain the imaging stability evaluation result of the optical microscope under test includes: When the imaging stability evaluation score of the optical microscope under test is outside the preset range and less than the minimum value within the preset range, the imaging stability evaluation result of the optical microscope under test is determined to be a decrease in imaging stability. Based on the environmental sensor data and operating status data of the optical microscope under test, the influencing factors affecting imaging stability are obtained, and an early warning signal and the influencing factors affecting imaging stability are issued. When the imaging stability evaluation score of the optical microscope under test is within the preset range, the imaging stability evaluation result of the optical microscope under test is judged to be passing. When the imaging stability evaluation score of the optical microscope under test is outside the preset range but higher than the maximum value within the preset range, the imaging stability evaluation result of the optical microscope under test is determined to be an increase in imaging stability.

10. An optical stability assessment and prediction system based on an improved LSTM model, characterized in that, The optical stability assessment and prediction method based on the improved LSTM model described in claim 1 is constructed, comprising: The data acquisition module is used to acquire imaging data, environmental sensor data, and operating status data of the optical microscope under test. The optical microscope imaging stability assessment module is used to: obtain the imaging stability assessment results of the optical microscope under test based on the imaging data, environmental sensor data and operating status data of the optical microscope under test, through the optical microscope imaging stability assessment model, wherein the optical microscope imaging stability assessment model is an improved LSTM model.

11. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the optical stability assessment and prediction method based on the improved LSTM model as described in any one of claims 1 to 9.

12. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the optical stability assessment and prediction method based on the improved LSTM model as described in any one of claims 1 to 9.

Images