Control method for automatically adjusting optical characteristics of target

Abstract

The invention provides a control method for automatically adjusting target optical characteristics, and relates to the technical field of optical characteristic adjustment, and the method comprises the steps: obtaining historical environment data, optical characteristic parameters during detection of optical equipment, and detection signal intensity to form an initial data set, and building an environment detection light beam model; modulating a main detection light beam according to real-time environment data in combination with an environment detection light beam model; acquiring detection echoes of environment noise and the main detection light beam; performing empirical mode decomposition and wavelet transform fusion processing on echo signals; extracting time-frequency domain characteristics of interference waves; interference waves are eliminated by predicting periodic interference characteristics and modulating optical characteristic parameters of an auxiliary detection light beam, and constructive interference with a main light beam is realized by adjusting optical characteristics of the auxiliary detection light beam for non-periodic interference waves, so that an enhanced main detection light beam is formed.

Description

Technical Field

[0001] The present invention relates to the technical field of optical property adjustment, and particularly to a control method for automatically adjusting target optical properties. Background Technique

[0002] With the continuous development of optical detection technology, how to accurately adjust and optimize optical properties in complex environments, especially in the presence of noise and interference waves, has become a major challenge faced by current technologies. When an optical detection device performs tasks, it often needs to adapt to different environmental changes, and these factors will affect the detection accuracy and signal quality of the optical device. Especially in the presence of environmental noise, periodic and aperiodic interference waves, the accuracy of the detected echo signal is often significantly affected, resulting in a decline in the performance of the detection system. Although some filtering and noise suppression methods are used in the prior art, there is a lack of an effective strategy for dynamically adjusting optical properties in a real-time environment, making it difficult to cope with the interference and noise generated under changing environmental conditions, thus affecting the stability and accuracy of the system.

[0003] In addition, traditional optical detection methods usually have difficulty distinguishing between periodic and aperiodic interference waves and cannot effectively make precise adjustments according to the characteristics of different interference waves. This results in the detection system being unable to automatically optimize and adjust in complex environments, especially for the elimination and enhancement of periodic and aperiodic interference waves. The prior art does not provide a unified solution. Therefore, how to automatically adjust optical properties according to real-time detection signals and effectively process different types of interference waves remains a technical problem to be solved urgently.

[0004] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure, and thus it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the present invention is to provide a control method for automatically adjusting target optical properties to solve the problems raised in the above background art.

[0006] To achieve the above purpose, the present invention provides the following technical solutions: A control method for automatically adjusting target optical properties, the specific steps including: Step 1: Obtain the historical environmental data of the optical detection area, the optical property parameters during the detection of the optical device, and the detection signal intensity to form an initial data set. After performing spatio-temporal alignment through Kalman filtering, establish an environmental detection beam model, and modulate the main detection beam according to the real-time environmental data in combination with the environmental detection beam model; Step 2: Obtain the environmental noise and the detection echo of the main detection beam. After performing empirical mode decomposition on the echo signal to obtain the intrinsic mode functions, perform wavelet transform fusion processing, strip the environmental noise, and extract the time-frequency domain characteristics of the interference wave; Step 3: According to the extracted time-frequency domain characteristics of the continuous interference wave, analyze the periodicity by calculating the autocorrelation function of the time-frequency domain characteristics of the interference wave, and determine whether the interference wave has a period; Step 4: For the periodic interference wave, establish a spatio-temporal prediction model, predict the periodic interference characteristics based on the environmental data, and perform interference cancellation by modulating the optical characteristic parameters of the auxiliary detection beam according to the predicted interference wave characteristics to eliminate the interference wave; Step 5: For the non-periodic interference wave, adjust the optical characteristics of the auxiliary detection beam to achieve constructive interference with the main beam, and form an enhanced main detection beam by increasing the optical characteristic parameters of the main detection beam.

[0007] Furthermore, the environmental data includes light intensity, temperature, humidity, and particulate matter concentration; The optical characteristic parameters include the wavelength, amplitude, phase, and frequency of the detection beam; The specific steps for noise reduction processing of the initial data set by the Kalman algorithm are as follows: State prediction equation: Among them, is the predicted state estimate at time represents the initial data set at the current moment, is the state transition matrix, is the control matrix of the optical characteristic parameters, is the control input; Covariance prediction equation: Among them, is the predicted error covariance matrix at time is the predicted error covariance matrix at time is the predicted error covariance matrix at time is the predicted error covariance matrix at time is the process noise covariance matrix, is the transpose matrix; Kalman gain calculation: Among them, is the Kalman gain, is the observation matrix of the detection signal intensity, is the observation transpose matrix of the detection signal intensity, is the measurement noise covariance matrix; State update equation: Among them, is the detection signal intensity at time , Is the corrected state estimate; Wherein, Is the moment The corrected error covariance matrix, Is the identity matrix.

[0008] Further, the steps of establishing the environmental detection beam model are: Establish an environmental detection beam model through a multi-layer perceptron. Train the model using the environmental data in the historical initial dataset as the input of the model, and the detection signal intensity and optical characteristic parameters as the output of the model to obtain a trained environmental detection beam model. Output the detection signal intensity and optical characteristic parameters according to the real-time environmental data. The expression form is: Wherein, Is the strongest detection signal intensity, Are respectively the amplitude, wavelength, phase, and frequency of the main detection beam, Is the multi-layer perceptron neural network model, Are the optical characteristic parameters of the main detection beam, Is the environmental data.

[0009] Further, after obtaining the intrinsic mode functions by performing empirical mode decomposition on the echo signal, the specific steps of the wavelet transform fusion processing are: The environmental noise And the detection echo signal of the main detection beam Are decomposed into multiple intrinsic mode functions and a residue through empirical mode decomposition; Wherein, Is the th intrinsic mode function of the environmental noise, Is the residue of the environmental noise, Is the th intrinsic mode function of the detection echo signal of the main detection beam, Is the residue of the detection echo signal of the main detection beam, Are respectively the numbers of the intrinsic mode functions of the environmental noise and the detection echo signal of the main detection beam, Are all positive integers.

[0010] Further, perform wavelet transform on each intrinsic mode function To obtain time-frequency domain features: Wherein, Is the time-frequency feature of the th intrinsic mode function of the environmental noise on the scale And position Of the wavelet basis function And position The scale of the th intrinsic mode function of the detection echo signal of the main detection beam on the wavelet basis function and the time-frequency characteristics at the position ; is the wavelet basis function, is the complex conjugate of the wavelet basis function, is the frequency scale parameter, is the time translation parameter; Among them, the calculation formula of the wavelet basis function is: Among them,

[0011] is the center frequency of the wavelet, is the time variable, represents the imaginary unit.

[0011] Furthermore, the calculation formula of the autocorrelation function of the time-frequency domain characteristics of the interference wave is: Among them, is the autocorrelation function of the time-frequency domain characteristics of the interference wave, is the time-frequency domain characteristics of the detection echo signal of the main detection beam at the moment ; is the time delay interval, represents the integral of the time variable ; According to the autocorrelation function generating periodic peaks at certain moments, judging the period of the interference wave: Among them, is the delay time of the second significant peak, is the delay time of the first significant peak; There is a period of the interference wave, judging that the interference wave has a period.

[0012] Furthermore, the steps for establishing the spatio-temporal prediction model are: Establishment of the spatio-temporal prediction model: The spatio-temporal prediction model is based on a long short-term memory neural network and includes an input gate, a forget gate, and an output gate; The equation expression of the forget gate is: Among them, is the current switch state of the forget gate, represents the sigmoid function, is the weight of the forget gate, is the bias parameter of the forget gate, is the hidden state at the previous moment, is the time-frequency domain feature of the interference wave in the current interference wave period; The equation expression of the input gate is: Among them, is the current switch state of the input gate, is the weight of the input gate, is the bias parameter of the input gate; Among them, candidate cell state, represents the hyperbolic tangent function, is the candidate cell weight, is the bias parameter of the candidate cell state; The update equation is: Among them, is the current cell state, is the cell state at the previous moment; The equation expression of the output gate is: Among them, is the current switch state of the output gate, is the weight of the output gate, is the bias parameter of the output gate, is the time-frequency domain feature of the interference wave in the next interference wave period; Among them, the mathematical expressions of the sigmoid function and the hyperbolic tangent function are: Among them, is the sigmoid function, is the input of the sigmoid function, is the hyperbolic tangent function; Training of the spatio-temporal prediction model: Using the time-frequency domain feature of the interference wave in the previous interference wave period in the historical data and the environmental data as the input of the model, and the time-frequency domain feature of the interference wave in the next interference wave period as the output of the model to train the model; Prediction of the spatio-temporal prediction model: Using the time-frequency domain feature of the interference wave in the current interference wave period and the environmental data as the input of the model to predict the time-frequency domain feature of the interference wave in the next interference wave period.

[0013] Furthermore, the steps of modulating the optical feature of the auxiliary detection beam according to the predicted interference wave feature are: Extracting the frequency, phase and amplitude of the interference wave according to the predicted time-frequency domain feature of the interference wave, and generating the interference wave according to the frequency, phase and amplitude: Among them, is the intensity of the interference wave at time , is the amplitude of the interference wave, is the frequency of the interference wave, is the phase of the interference wave; According to the interference cancellation principle, modulate the optical characteristic parameters of the auxiliary detection beam: When , , , it interferes with the interference wave; Among them, is the intensity of the auxiliary detection beam at time , is the amplitude of the auxiliary detection beam, is the frequency of the auxiliary detection beam, is the phase of the auxiliary detection beam.

[0014] Furthermore, the specific steps for adjusting the optical characteristics of the auxiliary detection beam are: The expression form of the main detection beam is: Among them, is the intensity of the main detection beam at time , are respectively the amplitude, phase, and frequency of the main detection beam; When achieving constructive interference with the main beam, it is necessary to satisfy , , ; The expression form of the enhanced main detection beam is: Among them, is the intensity of the enhanced main detection beam at time .

[0015] Compared with the prior art, the beneficial effects of the present invention are: The present invention forms an initial data set by obtaining historical environmental data, optical characteristic parameters during the detection of optical devices, and detection signal intensity, establishes an environmental detection beam model, modulates the main detection beam according to real-time environmental data combined with the environmental detection beam model, obtains the environmental noise and the detection echo of the main detection beam, performs empirical mode decomposition and wavelet transform fusion processing on the echo signal, extracts the time-frequency domain characteristics of the interference wave, calculates the periodicity of the time-frequency domain characteristics of the interference wave, for the periodic interference wave, predicts the periodic interference characteristics and modulates the optical characteristic parameters of the auxiliary detection beam to eliminate the interference wave, and for the non-periodic interference wave, adjusts the optical characteristics of the auxiliary detection beam to achieve constructive interference with the main beam and form an enhanced main detection beam Based on the introduction of technologies such as Kalman filtering, empirical mode decomposition, and wavelet transform, the present invention realizes the problem of automatically adjusting optical characteristics in complex environments and effectively coping with environmental noise and interference waves. First, through the spatio-temporal alignment of historical environmental data and optical characteristic parameters in the optical detection area, the characteristics of the main detection beam can be dynamically adjusted according to environmental changes to ensure its adaptation to real-time environmental conditions. This adjustment method can not only reduce errors caused by environmental fluctuations but also enhance the robustness of the system in uncertain environments.

[0016] The present invention also effectively extracts the time-frequency domain characteristics of interference waves by performing empirical mode decomposition and wavelet transform processing on echo signals, enabling accurate identification of periodic interference waves and non-periodic interference waves. For periodic interference waves, the system predicts their characteristics by establishing a spatio-temporal prediction model and eliminates the interference waves through adjusting the optical characteristics of the auxiliary detection beam for destructive interference; for non-periodic interference waves, the quality of the detection signal is enhanced through constructive interference by enhancing the optical characteristics of the main detection beam. Finally, through this method of automatically adjusting optical characteristics, the anti-interference ability and detection accuracy of the optical detection system in complex environments are significantly improved, ensuring the stability and efficiency of the system. It can optimize the performance of the optical detection system in real-time through the adjustment of optical characteristics in a dynamically changing environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the overall method flow of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] To make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below in conjunction with specific embodiments.

[0019] It should be noted that unless otherwise defined, the technical terms or scientific terms used in the present invention should have the ordinary meaning understood by those with ordinary skills in the field to which the present invention belongs. The "first", "second", and similar terms used in the present invention do not indicate any order, quantity, or importance, but are only used to distinguish different components. The terms such as "including" or "comprising" mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects. The terms such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The terms such as "upper", "lower", "left", "right", etc. are only used to represent relative positional relationships, and when the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0020] Embodiment

[0021] Please refer toFigure 1 , the present invention provides a technical solution: A control method for automatically adjusting target optical characteristics, the specific steps include: Step 1: Obtain the historical environmental data of the optical detection area, the optical characteristic parameters during the detection of the optical device, and the detection signal intensity to form an initial data set. After performing spatio-temporal alignment through Kalman filtering, establish an environmental detection beam model, and modulate the main detection beam according to the real-time environmental data combined with the environmental detection beam model.

[0022] The light intensity affects the amplitude of the light wave. Strong light causes the amplitude of the light wave to increase, while weak light causes it to decrease. In addition, changes in light intensity may cause changes in the wavelength distribution of the light source, especially in the infrared band, and the optical system for detection needs to adjust its wavelength and amplitude. Temperature changes affect the propagation speed of light and the refractive index of the medium, thereby causing changes in the wavelength, amplitude, and phase of light. A high-temperature environment may cause the refractive index to decrease, resulting in changes in the light beam, and further changing the phase and frequency of light. When the temperature is too high, a scattering effect may also be triggered, affecting the stability of the signal. Humidity has a particularly significant impact on optical characteristics, especially in the infrared band. When the humidity is high, the absorption effect of water vapor is significant, resulting in attenuation of the amplitude of light. Changes in humidity also cause fluctuations in the refractive index of the medium, thereby changing the phase of light. When the particulate concentration increases, strong scattering and absorption effects will be triggered, especially in the visible light and infrared light bands, resulting in attenuation of the amplitude of the detection signal and even changing the propagation path of light, thus affecting the phase and frequency response of the optical system. Therefore, according to the influence of environmental data on the propagation characteristics of light, the optical system is prompted to adjust the wavelength, amplitude, phase, and frequency of its detection beam to optimize the detection accuracy and stability.

[0023] Performing spatio-temporal alignment through Kalman filtering effectively improves the accuracy and stability of detection data, especially under complex and dynamic environmental conditions. Kalman filtering is a recursive algorithm that can estimate the state of the system in noisy data and effectively perform signal processing. By performing spatio-temporal alignment through Kalman filtering, the noise in environmental data and optical detection data can be effectively processed and removed. Especially during real-time detection, the measurement signal may be affected by various interferences, such as changes in light intensity and device errors. Through the spatio-temporal alignment of Kalman filtering, the historical data and real-time data can be smoothed, thereby improving the quality of the data and reducing the impact of noise on subsequent optical characteristic adjustments.

[0024] One of the core advantages of the Kalman filter is that it can perform spatio-temporal alignment of environmental data (such as light intensity, temperature, humidity, etc.) and detection signals (such as optical characteristic parameters, detection signal intensity, etc.) at multiple time points. In practical applications, the environmental factors and the responses of optical devices may not be completely synchronized. The Kalman filter can accurately perform spatio-temporal alignment of these data through recursive calculation and prediction, ensuring a high degree of consistency between real-time detection data and historical data.

[0025] The environmental data and optical characteristic parameters change over time. The Kalman filter can dynamically adjust the characteristics of the light beam through a prediction model to adapt to the changing environment. For example, when factors such as light intensity, temperature, and humidity change, the Kalman filter can update the environmental light beam model in real time to ensure that the modulation of the detection light beam always adapts to the environmental conditions and optimize the detection results.

[0026] In this embodiment, the environmental data includes light intensity, temperature, humidity, and particulate matter concentration; The optical characteristic parameters include the wavelength, amplitude, phase, and frequency of the detection light beam; The specific steps for noise reduction processing of the initial data set through the Kalman algorithm are as follows: State prediction equation: Where, is the predicted state estimate at time represents the initial data set at the current time, is the state transition matrix, is the control matrix of the optical characteristic parameters, is the control input; Covariance prediction equation: Where, is the predicted error covariance matrix at time is the predicted error covariance matrix at time is the predicted error covariance matrix at time is the predicted error covariance matrix at time is the process noise covariance matrix, is the transpose matrix; Kalman gain calculation: Where, is the Kalman gain, is the observation matrix of the detection signal intensity, is the observation transpose matrix of the detection signal intensity, is the measurement noise covariance matrix; State update equation: Where, is the detection signal intensity at time , is the corrected state estimate; Where, At time The corrected error covariance matrix, is the identity matrix.

[0027] An environmental detection beam model is established through a multi-layer perceptron, achieving efficient and flexible model fitting and feature extraction, thereby optimizing the environmental adaptability of the optical system. A multi-layer perceptron is a feedforward neural network commonly used for tasks such as classification, regression, and pattern recognition, with good non-linear modeling capabilities.

[0028] A multi-layer perceptron consists of multiple layers of neurons and can effectively capture and represent non-linear relationships in data through activation functions. In an optical detection system, the relationship between environmental factors (such as temperature, humidity, light, etc.) and the response of optical devices is usually highly non-linear. Traditional linear models often struggle to accurately describe such complex relationships, while a multi-layer perceptron can more accurately simulate the complex dependence between environmental data and the detection beam through multiple hidden layers and non-linear activation functions, providing a beam modulation model with higher accuracy. During the training process, a multi-layer perceptron can automatically learn useful features of the input data without the need for manual feature extraction. By training on historical environmental data and the detection signals of optical devices, a multi-layer perceptron can automatically identify the key factors in environmental data related to the characteristics of the detection beam. This ability to automatically learn features enables the model to adaptively adjust the beam characteristics according to different environmental conditions, thereby optimizing the detection results.

[0029] Environmental factors are usually highly dynamic and variable, involving the interaction of multiple variables. Through the hierarchical structure of a multi-layer perceptron, the model can effectively process and represent these complex interactions in multiple dimensions. Whether it is the influence of multiple factors, changing light conditions, or the combined effects of factors such as temperature, a multi-layer perceptron can effectively capture the relationships between these variables, thereby establishing a more accurate and robust environmental detection beam model.

[0030] In this embodiment, the steps for establishing the environmental detection beam model are as follows: An environmental detection beam model is established through a multi-layer perceptron. The model is trained using environmental data from the historical initial dataset as the input, and the detection signal intensity and optical characteristic parameters as the output, to obtain a trained environmental detection beam model. The detection signal intensity and optical characteristic parameters are output according to real-time environmental data, and the expression form is: where, is the strongest detection signal intensity, are respectively the amplitude, wavelength, phase, and frequency of the main detection beam, is the multi-layer perceptron neural network model, are the optical characteristic parameters of the main detection beam, is environmental data.

[0031] The environmental detection beam model is based on a multi-layer perceptron neural network model, including an input layer, a hidden layer, and an output layer; Input layer: Among them, is environmental data, is the th environmental data type in the environmental data, is the number of environmental data types; Hidden layer: Among them, is the intermediate value of the linear output of the hidden layer, are the weight matrix and threshold of the hidden layer respectively, is the activation function, Output layer: Among them, is the intermediate value of the linear output of the output layer, are the weight matrix and threshold of the output layer respectively, is the detection signal strength and optical characteristic parameter, is the softmax activation function, outputting probability values; Among them, the activation function is the ReLU activation function, and the specific formula is:

[0032] The specific formula of the Softmax activation function is: Among them, is the th input value of the Softmax activation function.

[0033] Step 2: Obtain the environmental noise and the detection echo of the main detection beam. After performing empirical mode decomposition on the echo signal to obtain the intrinsic mode functions, through wavelet transform fusion processing, by stripping the environmental noise, extract the time-frequency domain characteristics of the interference wave.

[0034] During the empirical mode decomposition process, the signal is decomposed into multiple intrinsic mode functions and a residual term. Each intrinsic mode function represents different frequency components of the signal.

[0035] Wavelet transform acts on these intrinsic mode functions to further analyze their time-frequency characteristics. By applying wavelet transform to each intrinsic mode function, the changes in the signal can be finely captured at multiple frequency bands and time scales, especially the local mutations or rapidly changing parts. This is very effective for the extraction of interference waves in the signal, noise removal, and in-depth analysis of time-frequency characteristics.

[0036] Wavelet transform is performed on each intrinsic mode function to identify the high-frequency components or mutation features therein. In this way, wavelet transform can reveal the details and sudden events in the signal. Wavelet transform can help strip off these high-frequency or interfering components while retaining the main features of the signal and optimizing the signal quality.

[0037] First, empirical mode decomposition is performed to decompose the signal into several intrinsic mode functions, which can separate different frequency components of the signal. Each intrinsic mode function can be regarded as different parts of the signal at different time scales. In particular, empirical mode can effectively process non-linear and non-stationary signals.

[0038] Then, wavelet transform is performed on each intrinsic mode function to further analyze its time-frequency characteristics and perform processing such as denoising and feature extraction. This method can strip off the noise and interference waves in the signal and observe the local features of the signal at multiple scales, especially optimizing in the time domain and frequency domain.

[0039] This method of combining empirical mode decomposition and wavelet transform makes full use of the advantages of both in non-stationary signal analysis, noise stripping, time-frequency feature extraction, etc. It enables the optical system to more efficiently and stably adjust the optical characteristics of the target in a complex and dynamic environment.

[0040] In this embodiment, after obtaining the intrinsic mode functions by performing empirical mode decomposition on the echo signal, the specific steps of the wavelet transform fusion processing are as follows: The environmental noise and the detection echo signal of the main detection beam are decomposed into multiple intrinsic mode functions and a residual term through empirical mode decomposition; Among them, is the th intrinsic mode function of the environmental noise, is the residual term of the environmental noise, is the th intrinsic mode function of the detection echo signal of the main detection beam, is the residual term of the detection echo signal of the main detection beam, are the numbers of the intrinsic mode functions of the environmental noise and the detection echo signal of the main detection beam respectively, both are positive integers.

[0041] In this embodiment, for each intrinsic mode function wavelet transform is performed to obtain time-frequency domain features: Among them, is the scale of the th intrinsic mode function of the environmental noise on the wavelet basis function ​ and position The time-frequency characteristics at are the th intrinsic mode function of the detection echo signal of the main detection beam on the wavelet basis function at the scale and position The time-frequency characteristics at is the wavelet basis function, is the complex conjugate of the wavelet basis function, is the scaling and translation of the wavelet basis function at the scale and position under is the frequency scale parameter, is the time translation parameter; Among them, the calculation formula of the wavelet basis function is: Among them, is the central frequency of the wavelet, is the time variable, represents the imaginary unit.

[0042] Step 3: According to the time-frequency domain characteristics of the extracted continuous interference wave, analyze the periodicity by calculating the autocorrelation function of the time-frequency domain characteristics of the interference wave, and judge whether the interference wave has a period.

[0043] The autocorrelation function is a tool commonly used to analyze the periodicity and repetitive characteristics of signals. In the present invention, using the autocorrelation function to analyze the time-frequency domain characteristics of the interference wave can effectively identify the periodicity of the interference wave, providing support for subsequent interference cancellation and optical adjustment. Periodic interference waves have fixed frequencies or periods. Therefore, the system can predict the appearance of periodic interference in advance and make preparations for optical adjustment in advance. Compared with non-periodic interference, the appearance of periodic interference waves is more regular and it is easy to eliminate or adjust the interference according to the spatio-temporal prediction model. Through precise periodic analysis and interference adjustment, it can be ensured that when the system faces periodic interference, it can take intervention measures in a timely and accurate manner to avoid the decline of the system performance.

[0044] In this embodiment, the calculation formula of the autocorrelation function of the time-frequency domain characteristics of the interference wave is: Among them, is the autocorrelation function of the time-frequency domain characteristics of the interference wave, is the time-frequency domain characteristics of the detection echo signal of the main detection beam at the moment , is the time delay interval, represents the integral of the time variable ; Judge the period of the interference wave according to the periodic peak generated by the autocorrelation function at some moments : Among them, is the delay time of the second significant peak, is the delay time of the first significant peak; There is a period of interference wave , and it is determined that the interference wave has a period.

[0045] Step 4: For the periodic interference wave, by establishing a spatio-temporal prediction model, predict the periodic interference characteristics based on the environmental data, and perform interference cancellation by modulating the optical characteristic parameters of the auxiliary detection beam to eliminate the interference wave.

[0046] The characteristics of the periodic interference wave (such as frequency, phase, etc.) have a certain regularity and can be accurately predicted through the spatio-temporal prediction model. Based on these predictions, the optical characteristics of the auxiliary detection beam (such as light intensity, phase, frequency, etc.) can be precisely adjusted to achieve interference cancellation with the interference wave. This adjustment ensures that the phase of the interference wave is correctly cancelled, achieving a higher cancellation effect. The spatio-temporal prediction model can adjust the optical characteristics in advance according to environmental changes, enabling the interference wave to be quickly eliminated when it appears. The prediction ability makes the adjustment more accurate and timely, avoiding lags or errors, ensuring the stability and accuracy of the optical characteristics, and ensuring the smooth progress of the optical measurement and detection process. It can also predict the change trend of the periodic interference wave based on real-time environmental data (such as temperature, humidity, air pressure, etc.). This enables the system to adaptively adjust the optical characteristics according to different environmental factors, ensuring the high efficiency and accuracy of interference wave cancellation under various external conditions.

[0047] In this embodiment, the establishment steps of the spatio-temporal prediction model are as follows: Establishment of the spatio-temporal prediction model: The spatio-temporal prediction model is based on a long short-term memory neural network and includes an input gate, a forget gate, and an output gate; The equation expression of the forget gate is: Among them, is the current switch state of the forget gate, represents the sigmoid function, is the weight of the forget gate, is the bias parameter of the forget gate, is the hidden state at the previous moment, is the time-frequency domain characteristic of the interference wave of the current interference wave period; The equation expression of the input gate is: Among them, is the current switch state of the input gate, is the weight of the input gate, is the bias parameter of the input gate; Among them, The candidate cell state represents the hyperbolic tangent function, is the candidate cell weight, is the candidate cell state bias parameter; The update equation is: Among them, is the current cell state, is the cell state at the previous moment; The equation expression of the output gate is: Among them, is the current switch state of the output gate, is the output gate weight, is the output gate bias parameter, is the time-frequency domain feature of the interference wave in the next interference wave period; Among them, the mathematical expressions of the sigmoid function and the hyperbolic tangent function are: Among them, is the sigmoid function, is the input of the sigmoid function, is the hyperbolic tangent function; Training of the spatio-temporal prediction model: Using the time-frequency domain features of the interference wave in the previous interference wave period in the historical data and the environmental data as the input of the model, and the time-frequency domain features of the interference wave in the next interference wave period as the output of the model, the model is trained; Prediction of the spatio-temporal prediction model: Using the time-frequency domain features of the interference wave in the current interference wave period and the environmental data as the input of the model, the time-frequency domain features of the interference wave in the next interference wave period are predicted.

[0048] The long short-term memory neural network can effectively process the time-series data of interference waves and capture long-term dependencies. The characteristics of interference waves, such as periodicity, amplitude variation, etc., usually have significant time-series properties. The long short-term memory neural network can learn this time-series characteristic and accurately predict the future behavior of interference waves. Compared with traditional neural networks, the long short-term memory neural network has memory units that can maintain long-term dependencies on historical data, and is particularly suitable for situations where the characteristics of interference waves exhibit periodic or complex non-linear changes. By analyzing environmental data in real time, the long short-term memory neural network can predict future waveforms based on historical interference wave characteristics and adjust the optical characteristics of the auxiliary detection beam according to these prediction information, so as to achieve the cancellation of interference waves or the enhancement of the main detection beam. The advantage of this modeling method is that it improves the accuracy and real-time performance of interference wave prediction, enables the system to respond more flexibly to different interference environments, enhances the anti-interference ability of the optical system, and at the same time avoids the limitations of traditional linear models in dealing with non-linear complex interference waves. The long short-term memory neural network not only improves the response speed of the system, but also maintains high stability and reliability in complex dynamic environments.

[0049] In this embodiment, the step of modulating the optical characteristics of the auxiliary detection beam according to the predicted interference wave characteristics is as follows: Extract the frequency, phase, and amplitude of the interference wave according to the predicted time-frequency domain characteristics of the interference wave, and generate the interference wave according to the frequency, phase, and amplitude: Among them, is the intensity of the interference wave at time , is the amplitude of the interference wave, is the frequency of the interference wave, is the phase of the interference wave; According to the interference cancellation principle, modulate the optical characteristic parameters of the auxiliary detection beam: When , , , it interferes with the interference wave; Among them, is the intensity of the auxiliary detection beam at time , is the amplitude of the auxiliary detection beam, is the frequency of the auxiliary detection beam, is the phase of the auxiliary detection beam.

[0050] The basic principle of interference cancellation is to cancel the interfering wave with the main beam or other beams through opposite phases. When optical characteristics (such as phase, frequency, amplitude, etc.) are precisely adjusted, the interference conditions can be optimized, enabling the interfering wave and the auxiliary detection beam to have opposite phases, thereby completely eliminating the interfering wave. The precise adjustment allows the system to perform customized optical adjustments for periodic interfering waves, maximizing the cancellation effect and avoiding the impact of the interfering wave on the target optical characteristics.

[0051] Adjusting optical characteristics not only helps to eliminate interfering waves but also improves the quality of the main detection beam. For example, in the case of periodic interfering waves, appropriately adjusting the optical characteristics of the auxiliary detection beam to interfere destructively with the interfering wave can avoid the influence of the interfering wave on the main beam. Using the modulation of the optical characteristics of the auxiliary beam for interference cancellation enables timely adjustment according to environmental changes and the periodicity of the interfering wave. By using a spatio-temporal prediction model to predict periodic interfering waves, the system can anticipate the interfering wave in advance and adjust the optical characteristics accordingly, quickly adjusting the interference conditions.

[0052] Step 5: For non-periodic interfering waves, by adjusting the optical characteristics of the auxiliary detection beam, constructive interference is achieved with the main beam, and by increasing the optical characteristic parameters of the main detection beam, the main detection beam is strengthened.

[0053] In this embodiment, the specific steps for adjusting the optical characteristics of the auxiliary detection beam are as follows: The expression form of the main detection beam is: where is the intensity of the main detection beam at time , are respectively the amplitude, phase, and frequency of the main detection beam; When achieving constructive interference with the main beam, it is necessary to satisfy , , ; The expression form of the strengthened main detection beam is: where is the intensity of the strengthened main detection beam at time .

[0054] When facing non-periodic interference waves, a method is adopted to achieve constructive interference with the main beam by adjusting the optical properties of the auxiliary detection beam. The purpose is to effectively suppress the interference waves by enhancing the intensity of the main detection beam. The core of this method lies in precisely controlling the optical properties of the auxiliary detection beam to generate appropriate adjustments in the phase and amplitude of the main beam, so that the two beams achieve constructive interference during the interference process, thereby effectively amplifying the light intensity of the main beam. The characteristics of non-periodic interference waves are that their frequency, amplitude and other characteristics are random or time-varying, and traditional methods for dealing with periodic interference waves may be difficult to cope with. Therefore, through this method of constructive interference, the brightness of the main beam can be increased, making the signal clearer. At the same time, the adjustment of the auxiliary beam avoids the influence of the interference waves on the main beam, thereby achieving the suppression of non-periodic interference waves and ensuring the accuracy of the detection signal. The advantage of adopting this method is that it can make flexible real-time adjustments for non-periodic interference waves, without relying on the fixed rules of the interference waves. By constructive interference, the main beam is strengthened, effectively improving the anti-interference ability and stability of the system, and ensuring the efficient operation of the system in a complex environment.

[0055] The above formulas are all dimensionless and take their numerical values for calculation. The formula is obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0056] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0057] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units. They can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0058] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all should be covered by the protection scope of this application.

Claims

1. A control method for automatically adjusting the optical properties of a target, characterized in that: The specific steps include: Step 1: Obtain the historical environmental data of the optical detection area, the optical characteristic parameters of the optical equipment during detection, and the detection signal strength to form an initial data set. After time-space alignment through Kalman filtering, establish an environmental detection beam model, and modulate the main detection beam based on the real-time environmental data combined with the environmental detection beam model; Step 2: Obtain the detection echo of the environmental noise and the main detection beam, perform empirical mode decomposition on the echo signal to obtain the intrinsic mode function, and then perform wavelet transform fusion processing to strip the environmental noise and extract the time-frequency domain characteristics of the interference wave; Step 3: According to the extracted time-frequency domain characteristics of the continuous interference wave, the periodicity is analyzed by calculating the autocorrelation function of the time-frequency domain characteristics of the interference wave to determine whether the interference wave has a period; Step 4: For periodic interference waves, a spatiotemporal prediction model is established to predict the periodic interference characteristics according to environmental data, and the optical characteristic parameters of the auxiliary detection beam are modulated to perform interference destructive according to the predicted interference wave characteristics to eliminate the interference waves; Step 5: For the non-periodic interference wave, the optical properties of the auxiliary detection beam are adjusted to achieve constructive interference with the main beam, and the optical property parameters of the main detection beam are increased to form an enhanced main detection beam.

2. A control method for automatically adjusting target optical properties according to claim 1, characterized in that: The environmental data include light intensity, temperature, humidity and particle concentration; The optical characteristic parameters include the wavelength, amplitude, phase, and frequency of the probe beam; The specific steps of denoising the initial data set using the Kalman algorithm are: State prediction equation: in, For the moment The predicted state estimate represents the initial data set at the current moment, is the state transfer matrix, is the control matrix of optical characteristic parameters, is the control input; Covariance prediction equation: in, It's time The predicted error covariance matrix, It's time The predicted error covariance matrix, is the process noise covariance matrix, is the transposed matrix; Kalman gain calculation: in, is the Kalman gain, is the observation matrix for detecting signal strength, is the observed transposed matrix of the detection signal strength, is the measurement noise covariance matrix; State update equation: in, For the moment The detection signal strength, is the revised state estimate; in, For the moment The corrected error covariance matrix, is the identity matrix.

3. A control method for automatically adjusting target optical properties according to claim 1, characterized in that: The steps of establishing the environment detection beam model are: The environment detection beam model is established through a multi-layer perceptron. The environment data in the historical initial data set is used as the input of the model, and the detection signal strength and optical characteristic parameters are used as the output of the model to train the model. The trained environment detection beam model is obtained, and the detection signal strength and optical characteristic parameters are output according to the real-time environment data. The expression is: in, is the strongest detection signal strength, are the amplitude, wavelength, phase and frequency of the main detection beam, is a multi-layer perceptron neural network model. is the optical characteristic parameter of the main detection beam, For environmental data.

4. A control method for automatically adjusting target optical properties according to claim 1, characterized in that: After the echo signal is subjected to empirical mode decomposition to obtain the intrinsic mode function, the specific steps of the wavelet transform fusion process are as follows: The environmental noise The detection echo signal of the main detection beam Through empirical mode decomposition into multiple intrinsic mode functions and a residual term; in, The environmental noise The intrinsic mode functions, is the residual term of the ambient noise, The detection echo signal of the main detection beam is The intrinsic mode functions, is the residual term of the probe echo signal of the main probe beam, are the number of intrinsic mode functions of the detection echo signal of the ambient noise and the main detection beam, All are positive integers.

5. A control method for automatically adjusting target optical properties according to claim 4, characterized in that: For each intrinsic mode function Wavelet transform, get the time-frequency domain features: in, The environmental noise The intrinsic mode function is in the wavelet basis function The scale on and location The time-frequency characteristics of The detection echo signal of the main detection beam is The intrinsic mode function is in the wavelet basis function The scale on and location The time-frequency characteristics of is the wavelet basis function, is the complex conjugate of the wavelet basis function, is the wavelet basis function at scale and location Zoom and pan under is the frequency scale parameter, is the time shift parameter; The calculation formula of wavelet basis function is: in, is the center frequency of the wavelet, is the time variable, Represents an imaginary unit.

6. The control method for automatically adjusting the optical properties of a target according to claim 1, characterized in that: The calculation formula of the autocorrelation function of the time-frequency domain characteristics of the interference wave is: in, is the autocorrelation function of the time-frequency domain characteristics of the interference wave, The detection echo signal of the main detection beam at time The time-frequency domain characteristics of is the time delay interval, Represents the time variable Perform integration; According to the autocorrelation function Generate periodic peaks at certain moments to determine the period of the interference wave : in, is the delay time of the second significant peak, is the delay time of the first significant peak; There is a period of interference wave , determine whether the interference wave has a period.

7. The control method for automatically adjusting the optical properties of a target according to claim 1, characterized in that: The steps for establishing the spatiotemporal prediction model are: Establishment of spatiotemporal prediction model: The spatiotemporal prediction model is based on a long short-term memory neural network and includes an input gate, a forget gate, and an output gate; The equation of the forget gate is: in, is the current switch state of the forget gate, represents the sigmoid function, is the forget gate weight, is the forget gate bias parameter, is the hidden state at the previous moment, is the time-frequency domain characteristics of the interference wave in the current interference wave period; The equation for the input gate is: in, is the current switch state of the input gate, is the input gate weight, is the input gate bias parameter; in, Candidate cell states, To represent the hyperbolic tangent function, is the candidate cell weight, is the candidate cell state bias parameter; The update equation is: in, is the current cell state, is the cell state at the previous moment; The output gate equation is: in, is the current switch state of the output gate, is the output gate weight, is the output gate bias parameter, is the time-frequency domain characteristics of the interference wave of the next interference wave period; Among them, the mathematical expressions of the sigmoid function and the hyperbolic tangent function are: in, is the sigmoid function, is the input of the sigmoid function, is the hyperbolic tangent function; Training of the spatiotemporal prediction model: The model is trained based on the time-frequency domain characteristics of the interference wave in the previous interference wave cycle in the historical data and the environmental data as the input of the model, and the time-frequency domain characteristics of the interference wave in the next interference wave cycle as the output of the model; Prediction of the spatiotemporal prediction model: Based on the time-frequency domain characteristics of the interference wave in the current interference wave period and the environmental data as the input of the model, the time-frequency domain characteristics of the interference wave in the next interference wave period are predicted.

8. The control method for automatically adjusting the optical properties of a target according to claim 1, characterized in that: The steps of modulating the optical characteristics of the auxiliary detection beam according to the predicted interference wave characteristics are: The frequency, phase and amplitude of the interference wave are extracted according to the predicted time-frequency domain characteristics of the interference wave, and the interference wave is generated according to the frequency, phase and amplitude: in, The interference wave at time The strength of is the amplitude of the interference wave, is the frequency of the interference wave, is the phase of the interference wave; According to the principle of destructive interference, the optical characteristic parameters of the auxiliary detection beam are modulated: when , , When , it interferes with the interference wave; in, The auxiliary detection beam is at time The strength of is the amplitude of the auxiliary probe beam, is the frequency of the auxiliary probe beam, The phase of the auxiliary detection beam.

9. The control method for automatically adjusting the optical properties of a target according to claim 1, characterized in that: The specific steps for adjusting the optical properties of the auxiliary detection beam are: The main detection beam is expressed as: in, is the intensity of the main detection beam at time, are the amplitude, phase, and frequency of the main probe beam, respectively; When constructive interference is achieved with the main beam, , , ; The expression of the enhanced main detection beam is: in, To strengthen the main detection beam at time strength.

Images