A signal noise filtering method, device, storage medium and laser radar

Abstract

A signal noise filtering method, apparatus, storage medium, and lidar are disclosed. The method includes: performing ensemble empirical mode decomposition on an initial difference frequency signal generated by the lidar to obtain a component set corresponding to the initial difference frequency signal (S101); obtaining the autocorrelation function energy value corresponding to each noisy component in the component set, and obtaining the boundary component corresponding to the largest autocorrelation function energy value among the noisy components (S102); performing wavelet threshold denoising processing on the adjacent higher-order noisy components of the boundary component to obtain the denoised component corresponding to the adjacent higher-order noisy component (S103); and performing signal reconstruction processing based on the frequency band region of the initial difference frequency signal in the spectrum, and based on the denoised component and the boundary component, to obtain the denoised time-domain difference frequency signal (S104). This method can improve the signal-to-noise ratio of the difference frequency signal and increase the success rate of effective difference frequency extraction.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a signal noise filtering method, apparatus, storage medium, and lidar. Background Technology

[0002] Frequency Modulated Continuous Wave (FMCW) lidar is a type of continuous wave lidar based on coherent detection. It transmits a linearly varying frequency continuous wave as the transmitted signal during a frequency sweep period. A portion of the transmitted signal serves as the local oscillator signal, while the remainder is emitted for detection. The echo signal reflected from the object and returned forms a difference frequency signal with the local oscillator signal. However, in actual detection, the signal is easily affected by inherent noise from the lidar system and the environment, resulting in a low signal-to-noise ratio and making it difficult to extract the effective difference frequency signal. Summary of the Invention

[0003] This application provides a signal noise filtering method, apparatus, storage medium, and lidar, which can improve the signal-to-noise ratio of difference frequency signals and increase the success rate of effective difference frequency extraction.

[0004] One embodiment of this application provides a signal noise filtering method, including:

[0005] The initial difference frequency signal generated by the lidar is subjected to ensemble empirical mode decomposition to obtain the component set corresponding to the initial difference frequency signal;

[0006] Obtain the autocorrelation function energy value corresponding to each noisy component in the component set, and obtain the boundary component corresponding to the largest autocorrelation function energy value among the noisy components;

[0007] Wavelet threshold denoising is performed on the adjacent higher-order noisy components of the boundary component to obtain the denoised components corresponding to the adjacent higher-order noisy components. The adjacent higher-order noisy components are the noisy components in the component set that are adjacent to the boundary component and have a frequency fluctuation range greater than that of the boundary component.

[0008] Based on the frequency band region of the initial difference frequency signal in the spectrum, and based on the denoised component and the boundary component, signal reconstruction processing is performed to obtain the denoised time-domain difference frequency signal.

[0009] One embodiment of this application provides a signal noise filtering device, including:

[0010] The component set acquisition unit is used to perform ensemble empirical mode decomposition on the initial difference frequency signal generated by the lidar to obtain the component set corresponding to the initial difference frequency signal.

[0011] The boundary component acquisition unit is used to acquire the autocorrelation function energy value corresponding to each noisy component in the component set, and to acquire the boundary component corresponding to the largest autocorrelation function energy value among the noisy components.

[0012] The denoising component acquisition unit is used to perform wavelet threshold denoising processing on the adjacent higher-order noisy components of the boundary component to obtain the denoised components corresponding to the adjacent higher-order noisy components. The adjacent higher-order noisy components are the noisy components in the component set that are adjacent to the boundary component and have a frequency fluctuation range greater than that of the boundary component.

[0013] The signal reconstruction unit is used to perform signal reconstruction processing based on the frequency band region of the initial difference frequency signal in the spectrum, and based on the denoised component and the boundary component, to obtain the denoised time-domain difference frequency signal.

[0014] One embodiment of this application provides a computer storage medium storing a computer program, the computer program including program instructions, which, when executed by a processor, perform the above-described method steps.

[0015] One embodiment of this application provides a lidar, including a processor, a memory, and an input / output interface;

[0016] The processor is connected to the memory and the input / output interface, respectively. The input / output interface is used for page interaction, the memory is used to store program code, and the processor is used to call the program code to execute the above-described method steps.

[0017] In this embodiment, by performing component decomposition on the initial difference frequency signal of the lidar, the autocorrelation function energy values ​​of each noisy component in the component set are obtained. Based on the autocorrelation function energy values, the dominant boundary component in the component set can be determined. Since the adjacent higher-order components of the boundary component are mainly noise, wavelet thresholding is used to denoise the adjacent higher-order components of the boundary component, resulting in denoised components corresponding to the adjacent higher-order components. Finally, signal reconstruction processing is performed on the denoised components and the boundary components to obtain the denoised time-domain difference frequency signal. By combining adaptive mode decomposition and wavelet thresholding to achieve the noise filtering process of the difference frequency signal, the signal-to-noise ratio of the difference frequency signal is improved, thereby increasing the success rate of effective difference frequency extraction. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application 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 only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a system architecture diagram of signal and noise filtering provided in the embodiments of this application;

[0020] Figure 2 This is a schematic flowchart of a signal noise filtering method provided in an embodiment of this application;

[0021] Figure 3 This is a schematic flowchart of a signal noise filtering method provided in an embodiment of this application;

[0022] Figure 4 This is a schematic diagram illustrating an example of set empirical mode decomposition provided in an embodiment of this application;

[0023] Figure 5 This is a schematic diagram of an energy collection generation process provided in an embodiment of this application;

[0024] Figure 6 This is a schematic diagram of a process for determining boundary components provided in an embodiment of this application;

[0025] Figure 7 This is an example schematic diagram of an autocorrelation function energy curve provided in an embodiment of this application;

[0026] Figure 8 This is an example schematic diagram of an autocorrelation function energy curve provided in an embodiment of this application;

[0027] Figure 9 This is a schematic diagram of the structure of a signal noise filtering device provided in an embodiment of this application;

[0028] Figure 10 This is a schematic diagram of the structure of a signal noise filtering device provided in an embodiment of this application;

[0029] Figure 11 This is a schematic diagram of the boundary component acquisition unit provided in an embodiment of this application;

[0030] Figure 12 This is a schematic diagram of the structure of a lidar provided in an embodiment of this application. Detailed Implementation

[0031] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0032] Please combine Figures 1-8 The illustrated embodiment provides a detailed description of the signal and noise filtering method provided in this application.

[0033] See Figure 1 This application provides a system architecture diagram for signal and noise filtering. For example... Figure 1 As shown, the embodiments of this application can be applied to scenarios of lidar detection, such as environmental monitoring, aerospace, communication, autonomous driving navigation, positioning and other detection scenarios. The lidar's transmitted signal changes periodically according to the law of triangular waves to transmit signals to the detection target and receive the echo signal returned by the detection target. The initial difference frequency signal formed by the transmitted signal and the echo signal is obtained. The initial difference frequency signal can be processed by a signal processor through a series of signal processing processes, including analog-to-digital conversion processing, signal filtering processing, signal data extraction, signal data calculation, etc. Then, the signal spectrum and data generated by the signal processor are stored, displayed and managed by the background management device.

[0034] Because the transmitted and echo signals are easily affected by inherent noise from the lidar system and the environment, they present a noisy initial difference frequency signal in the spectrum. To remove this noise, this application proposes a signal noise filtering device. This device can be integrated into the signal processor or used as a standalone device to filter noise from the initial difference frequency signal. The device performs ensemble empirical mode decomposition on the initial difference frequency signal generated by the lidar to obtain a set of components corresponding to the initial difference frequency signal. The signal noise filtering device then acquires the components from the set of components... The autocorrelation function energy value corresponding to the noisy component is used to obtain the boundary component corresponding to the largest autocorrelation function energy value among all noisy components. The signal noise filtering device performs wavelet threshold denoising processing on the adjacent higher-order noisy components of the boundary component to obtain the denoised component corresponding to the adjacent higher-order noisy component. The adjacent higher-order noisy component is the noisy component that is adjacent to the boundary component in the component set and has a frequency fluctuation range greater than the boundary component. The signal noise filtering device performs signal reconstruction processing based on the frequency band region of the initial difference frequency signal in the spectrum and based on the denoised component and the boundary component to obtain the denoised time-domain difference frequency signal. By performing component decomposition on the initial difference frequency signal of the lidar, the autocorrelation function energy values ​​of each noisy component in the component set are obtained. Based on the autocorrelation function energy values, the dominant boundary component in the component set can be determined. Since the adjacent higher-order components of the boundary component are mainly noise, wavelet thresholding is used to denoise the adjacent higher-order components of the boundary component, obtaining the denoised components corresponding to the adjacent higher-order components. Finally, signal reconstruction processing is performed on the denoised components and the boundary components to obtain the denoised time-domain difference frequency signal. By combining adaptive mode decomposition and wavelet thresholding to achieve the noise filtering process of the difference frequency signal, the signal-to-noise ratio of the difference frequency signal is improved, thereby increasing the success rate of effective difference frequency extraction.

[0035] based on Figure 1 For the system architecture, please refer to [link / reference]. Figure 2 This is a flowchart illustrating a signal-noise filtering method provided in an embodiment of this application. Figure 2 As shown, the method described in this application embodiment may include the following steps S101-S104.

[0036] S101, Perform ensemble empirical mode decomposition on the initial difference frequency signal generated by the lidar to obtain the component set corresponding to the initial difference frequency signal;

[0037] Specifically, the signal noise filtering device performs ensemble empirical mode decomposition (EEMD) on the initial difference frequency signal generated by the lidar. The initial difference frequency signal can be a difference frequency signal containing noise, that is, the difference frequency signal of the lidar for detecting the target without signal processing. After EEMD processing, the component set corresponding to the initial difference frequency signal can be obtained. The component set includes multiple noisy components, which can include multiple intrinsic mode components and a residual component. The intrinsic mode components and the residual component are arranged in order of frequency fluctuation range, which can represent the signal frequency range in the frequency domain of the noisy component.

[0038] S102, obtain the autocorrelation function energy value corresponding to each noisy component in the component set, and obtain the boundary component corresponding to the largest autocorrelation function energy value among the noisy components;

[0039] Specifically, the signal noise filtering device can calculate the energy value of the autocorrelation function corresponding to each noisy component in the component set based on the autocorrelation function corresponding to each noisy component in the component set. The autocorrelation function can be an unbiased autocorrelation function, which reflects the correlation between the values ​​of the signal represented by the noisy component at any two different times. The signal noise filtering device can first calculate the autocorrelation function of each noisy component, and then generate the energy set of the initial difference frequency signal based on the autocorrelation function. The energy set can specifically include the energy value of the autocorrelation function corresponding to each noisy component in the component set. The signal noise filtering device can obtain the maximum autocorrelation function energy value in the energy set and determine the noisy component corresponding to the maximum autocorrelation function energy value as the boundary component. The boundary component can specifically be the noisy component dominated by the useful signal in the initial difference frequency signal. The useful signal specifically represents the real and effective difference frequency signal returned by the detected target after the transmitted signal.

[0040] S103, perform wavelet threshold denoising on the adjacent higher-order noisy components of the boundary component to obtain the denoised component corresponding to the adjacent higher-order noisy component.

[0041] Specifically, the signal noise filtering device can perform wavelet threshold denoising on the adjacent higher-order noisy components of the boundary component to obtain the denoised components corresponding to the adjacent higher-order noisy components. The adjacent higher-order noisy components are the noisy components in the component set that are adjacent to the boundary component and have a frequency fluctuation range greater than the boundary component, that is, the noisy components that are one order higher than the boundary component. It can be understood that since high frequencies in a signal are mainly noise, when converted to the wavelet domain, they are mainly represented by high-frequency coefficients. Therefore, by performing wavelet threshold denoising, the high-frequency coefficients representing noise in the wavelet domain can be set to zero or contracted to achieve the purpose of denoising.

[0042] S104, based on the frequency band region of the initial difference frequency signal in the spectrum, and based on the denoised component and the boundary component, perform signal reconstruction processing to obtain the denoised time-domain difference frequency signal.

[0043] Specifically, the signal noise filtering device can perform signal reconstruction processing based on the frequency band region of the initial difference frequency signal in the spectrum, and based on the denoised component and the boundary component, to obtain the denoised time-domain difference frequency signal. The frequency band region can be divided into high-frequency region, mid-frequency region and low-frequency region according to different frequency band thresholds. The frequency band thresholds for the frequency band region division can be set according to the actual situation. The frequency band region can be specifically represented as a frequency division range. The signal noise filtering device can obtain the frequency value of the initial difference frequency signal and obtain the frequency band region where the frequency value is located. The signal noise filtering device can obtain the signal reconstruction method corresponding to the frequency band region, and perform signal reconstruction processing on the denoised component and the boundary component based on the signal reconstruction method to obtain the denoised time-domain difference frequency signal.

[0044] It should be noted that both the time-domain difference frequency signal and the initial difference frequency signal can be represented as difference frequency signals in the time domain. The initial difference frequency signal is the difference frequency signal in the time domain before denoising, and the time-domain difference frequency signal is the difference frequency signal in the time domain after denoising.

[0045] In this embodiment, by performing component decomposition on the initial difference frequency signal of the lidar, the autocorrelation function energy values ​​of each noisy component in the component set are obtained. Based on the autocorrelation function energy values, the dominant boundary component in the component set can be determined. Since the adjacent higher-order components of the boundary component are mainly noise, wavelet thresholding is used to denoise the adjacent higher-order components of the boundary component, resulting in denoised components corresponding to the adjacent higher-order components. Finally, signal reconstruction processing is performed on the denoised components and the boundary components to obtain the denoised time-domain difference frequency signal. By combining adaptive mode decomposition and wavelet thresholding to achieve the noise filtering process of the difference frequency signal, the signal-to-noise ratio of the difference frequency signal is improved, thereby increasing the success rate of effective difference frequency extraction.

[0046] based on Figure 1 For the system architecture, please refer to [link / reference]. Figure 3 This is a flowchart illustrating a signal-noise filtering method provided in an embodiment of this application. Figure 3 As shown, the method described in this application embodiment may include the following steps S201-S206.

[0047] S201, Perform ensemble empirical mode decomposition on the initial difference frequency signal generated by the lidar to obtain the component set corresponding to the initial difference frequency signal;

[0048] Specifically, the signal noise filtering device performs EEMD processing on the initial difference frequency signal generated by the lidar. The initial difference frequency signal can be a difference frequency signal containing noise, that is, the difference frequency signal of the lidar for detecting the target without signal processing. After EEMD processing, the component set corresponding to the initial difference frequency signal can be obtained. The component set includes multiple noisy components. The multiple noisy components can include multiple intrinsic mode components and a residual component. The intrinsic mode components and the residual component are arranged in order of frequency fluctuation range. The frequency fluctuation range can represent the signal frequency range in the frequency domain where the noisy component is located.

[0049] Optionally, assuming the initial difference frequency signal is x(t), then performing EEMD processing on x(t) can yield m intrinsic mode components c. i (t) and a residual component r(t).

[0050]

[0051] Where m represents the number of intrinsic modal components, t represents the time of the component, and i represents the i-th noisy component, where i is less than or equal to m. In this application, both the intrinsic modal components and residual components can be noisy components included in the component set. For example, please refer to [reference needed]. Figure 4 ,like Figure 4As shown, x represents the initial difference frequency signal. Assume that EEMD processing of x yields eight intrinsic mode components (IMF1-IMF8) and a residual component r. IMF1-IMF8 represent the first to eighth order noisy components, respectively. The frequency fluctuation range of the first order noisy component is 0.15f~0.5f, the second order noisy component is 0.05f~0.25f, the third order noisy component is 0.03f~0.13f, the fourth order noisy component is 0.02f~0.075f, the fifth order noisy component is 0.01f~0.03f, the sixth order noisy component is 0.01f~0.025f, the seventh order noisy component is 0~0.02f, the eighth order noisy component is 0~0.015f, and the residual component is 0~0.01f. Although the frequency fluctuation ranges of adjacent noisy components partially overlap, by comparing one or more of the maximum, minimum, average, and median frequency values ​​of the frequency fluctuation ranges of multiple noisy components, it can be seen that the frequency fluctuation range from the first-order noisy component to the residual component shows a decreasing trend.

[0052] S202, obtain the autocorrelation function corresponding to each noisy component in the component set, and generate the energy set of the initial difference frequency signal based on the autocorrelation function;

[0053] Specifically, the signal noise filtering device can obtain the autocorrelation function corresponding to each noisy component in the component set, and generate the energy set of the initial difference frequency signal based on the autocorrelation function. The energy set includes the energy value of the autocorrelation function corresponding to each noisy component in the component set. The autocorrelation function can be an unbiased autocorrelation function, which reflects the correlation between the values ​​of the signal represented by the noisy component at any two different times. Optionally, the signal noise filtering device can obtain any two component values ​​of the target noisy component in the component set. The target noisy component is any noisy component in the component set, and the component values ​​are the component values ​​corresponding to any two times of the target noisy component. The signal noise filtering device can calculate the autocorrelation function of the target noisy component based on the component values. The autocorrelation function can be expressed by the following formula:

[0054] Ri ( t 1, t 2) = E [ ci ( t 1) ci ( t 2)]

[0055] Where c represents any noisy component in the component set, i.e., the target noisy component, and t1 and t2 represent two arbitrary times in the target noisy component.

[0056] The signal-noise filtering device can calculate the autocorrelation function energy value of the target noisy component based on the autocorrelation function. The autocorrelation function energy value can be calculated using the following formula:

[0057]

[0058] Where i represents the i-th noisy component in the component set.

[0059] The signal noise filtering device can add the autocorrelation function energy value of the target noisy component to the energy set of the initial difference frequency signal. Similarly, for the remaining components in the component set, the corresponding autocorrelation function energy value can be obtained according to the above calculation process of the target noisy component and added to the energy set. The energy set may include the autocorrelation function energy value corresponding to each noisy component in the component set.

[0060] S203, obtain the maximum autocorrelation function energy value in the energy set, and determine the noisy component corresponding to the maximum autocorrelation function energy value as the boundary component;

[0061] Specifically, the signal noise filtering device can obtain the maximum autocorrelation function energy value in the energy set and determine the noisy component corresponding to the maximum autocorrelation function energy value as the boundary component. Optionally, the signal noise filtering device can generate an autocorrelation function energy curve based on the respective autocorrelation function energy values ​​in the energy set. The signal noise filtering device can obtain the maximum autocorrelation function energy value in the autocorrelation function energy curve and determine the noisy component corresponding to the maximum autocorrelation function energy value as the boundary component. That is, it can quickly and accurately obtain the noisy component dominated by the useful signal in the initial difference frequency signal. The useful signal specifically represents the real and effective difference frequency signal returned by the transmitted signal after passing through the detection target. For example: when the maximum value of the autocorrelation function energy in the autocorrelation function energy curve is located on the first-order noisy component, that is, the first-order noisy component in the component set is the noisy component dominated by the useful signal in the initial difference frequency signal, the first-order noisy component is determined as the boundary component; when the maximum value of the autocorrelation function energy in the autocorrelation function energy curve is located on the k-th noisy component, that is, the k-th noisy component in the component set is the noisy component dominated by the useful signal in the initial difference frequency signal, the k-th noisy component is determined as the boundary component, where k is a positive integer greater than 1.

[0062] S204, perform wavelet threshold denoising on the adjacent higher-order noisy components of the boundary component to obtain the denoised components corresponding to the adjacent higher-order noisy components.

[0063] Specifically, the signal noise filtering device can perform wavelet threshold denoising on the adjacent higher-order noisy components of the boundary component to obtain the denoised components corresponding to the adjacent higher-order noisy components. The adjacent higher-order noisy components are the noisy components in the component set that are adjacent to the boundary component and have a frequency fluctuation range greater than the boundary component, that is, the noisy components that are one order higher than the boundary component. It can be understood that since high frequencies in a signal are mainly noise, when converted to the wavelet domain, they are mainly represented by high-frequency coefficients. Therefore, by performing wavelet threshold denoising, the high-frequency coefficients representing noise in the wavelet domain can be set to zero or contracted to achieve the purpose of denoising.

[0064] It should be noted that when the boundary component is a first-order noisy component in the component set, wavelet threshold denoising is performed on the boundary component to obtain the first denoised component corresponding to the boundary component; when the boundary component is a non-first-order noisy component (e.g., the kth order) in the component set, wavelet threshold denoising is performed on the adjacent higher-order noisy components (e.g., the (k-1th order) of the boundary component to obtain the second denoised component corresponding to the adjacent higher-order noisy component.

[0065] S205, based on the frequency band region of the initial difference frequency signal in the spectrum, and based on the denoised component and the boundary component, signal reconstruction processing is performed to obtain the denoised time-domain difference frequency signal;

[0066] Specifically, the signal noise filtering device can perform signal reconstruction processing based on the frequency band region of the initial difference frequency signal in the spectrum, and based on the denoised component and the boundary component, to obtain the denoised time-domain difference frequency signal. The frequency band region can be divided into high-frequency region, mid-frequency region and low-frequency region according to different frequency band thresholds. The frequency band thresholds for the frequency band region division can be set according to the actual situation. The frequency band region can be specifically represented as a frequency division range. The signal noise filtering device can obtain the frequency value of the initial difference frequency signal and obtain the frequency band region where the frequency value is located. The signal noise filtering device can obtain the signal reconstruction method corresponding to the frequency band region, and perform signal reconstruction processing on the denoised component and the boundary component based on the signal reconstruction method to obtain the denoised time-domain difference frequency signal.

[0067] Optionally, in a first feasible embodiment of this application, when the boundary component is a first-order noisy component in the component set, the signal noise filtering process can be performed as follows for signal reconstruction.

[0068] (1) When the initial difference frequency signal is in the first frequency band region in the spectrum, the first denoised component and the second-order noisy component in the component set are subjected to signal reconstruction processing to obtain the denoised time-domain difference frequency signal. Specifically, the following signal reconstruction method can be adopted:

[0069] x'(t) = c'(1) + c(2)

[0070] Where x'(t) represents the time-domain difference frequency signal after denoising, c'(1) represents the first denoised component corresponding to the first-order noisy component, and c(2) represents the second-order noisy component in the component set.

[0071] (2) When the initial difference frequency signal is in the second frequency band region in the spectrum, the first denoised component, the second-order noisy component and the third-order noisy component in the component set are subjected to signal reconstruction processing to obtain the denoised time-domain difference frequency signal. Specifically, the following signal reconstruction methods can be adopted:

[0072] x'(t)=c' (1) + c (2) + c (3)

[0073] Where x'(t) represents the time-domain difference frequency signal after denoising, c'(1) represents the first denoised component corresponding to the first-order noisy component, c(2) represents the second-order noisy component in the component set, and c(3) represents the third-order noisy component in the component set.

[0074] (3) When the initial difference frequency signal is in the third frequency band region in the spectrum, the first denoised component and the remaining noisy component are subjected to signal reconstruction processing to obtain the denoised time-domain difference frequency signal. The remaining noisy component is the other noisy component in the component set except for the first-order noisy component. Specifically, the following signal reconstruction method can be adopted:

[0075] x'(t)=c' (1) + c (2) +…+ r (t)

[0076] Where x'(t) represents the time-domain difference frequency signal after denoising, c'(1) represents the first denoised component corresponding to the first-order noisy component, c(2) represents the second-order noisy component in the component set, and r(t) represents the last-order noisy component in the component set, i.e. the residual component.

[0077] In a second feasible embodiment of this application, when the boundary component is a non-first-order noisy component in the component set, taking the kth order as an example, the signal noise filtering process can be performed by signal reconstruction in the following manner.

[0078] (1) When the initial difference frequency signal is in the first frequency band region in the spectrum, the second denoised component and the boundary component are subjected to signal reconstruction processing to obtain the denoised time-domain difference frequency signal. Specifically, the following signal reconstruction method can be adopted:

[0079] x'(t) = c'(k-1) + c(k)

[0080] Where x'(t) represents the time-domain difference frequency signal after denoising, c'(k-1) represents the second denoised component corresponding to the (k-1)th order noisy component, and c(k) represents the kth order noisy component in the component set.

[0081] (2) When the initial difference frequency signal is in the second frequency band region in the spectrum, the second denoised component, the boundary component, and the adjacent low-order noisy components of the boundary component are subjected to signal reconstruction processing to obtain the denoised time-domain difference frequency signal. The adjacent low-order noisy components are the noisy components in the component set that are adjacent to the boundary component and have a frequency fluctuation range smaller than the boundary component, that is, the noisy components that are one order lower than the boundary component. Specifically, the following signal reconstruction methods can be adopted:

[0082] x'(t) = c'(k-1) + c(k) + c(k+1)

[0083] Where x'(t) represents the time-domain difference frequency signal after denoising, c'(k-1) represents the second denoised component corresponding to the (k-1)th order noisy component, c(k) represents the kth order noisy component in the component set, and c'(k+1) represents the (k+1)th order noisy component.

[0084] (3) When the initial difference frequency signal is in the third frequency band region in the spectrum, the second denoised component, the boundary component, and the remaining low-order noisy components of the boundary component are subjected to signal reconstruction processing to obtain the denoised time-domain difference frequency signal. The remaining low-order noisy components are all noisy components in the component set whose frequency fluctuation range is smaller than that of the boundary component. Specifically, the following signal reconstruction methods can be adopted:

[0085] x'(t)=c'(k-1) + c(k) +…+ r(t)

[0086] Where x'(t) represents the time-domain difference frequency signal after denoising, c'(k-1) represents the second denoised component corresponding to the (k-1)th order noisy component, c(k) represents the kth order noisy component in the component set, and r(t) represents the last order noisy component in the component set, i.e. the residual component.

[0087] It should be noted that the maximum frequency value of the second frequency band region is less than the minimum frequency value of the first frequency band region, and the minimum frequency value of the second frequency band region is greater than the maximum frequency value of the third frequency band region. That is, the first frequency band region represents the high-frequency region, the second frequency band region represents the mid-frequency region, and the third frequency band region represents the low-frequency region. Both the time-domain difference frequency signal and the initial difference frequency signal can be represented as difference frequency signals in the time domain. The initial difference frequency signal is the difference frequency signal in the time domain before denoising, and the time-domain difference frequency signal is the difference frequency signal in the time domain after denoising.

[0088] S206, Perform a fast Fourier transform on the time-domain difference frequency signal to obtain a frequency-domain difference frequency signal, and obtain the difference frequency value corresponding to the maximum amplitude value in the frequency-domain difference frequency signal.

[0089] Specifically, the signal noise filtering device can perform a fast Fourier transform on the time-domain difference frequency signal to obtain a frequency-domain difference frequency signal. The difference frequency value corresponding to the maximum amplitude value is obtained in the frequency-domain difference frequency signal. The frequency-domain difference frequency signal can be specifically represented as the difference frequency signal in the frequency domain after noise reduction. The signal noise filtering device can obtain the position of the maximum amplitude value in the spectrum diagram formed by the frequency-domain difference frequency signal and determine the frequency value corresponding to the position as the difference frequency value of the useful signal. The useful signal can be specifically represented as the real and effective difference frequency signal returned by the transmitted signal after passing through the detection target.

[0090] In this embodiment, by performing component decomposition on the initial difference frequency signal of the lidar, the autocorrelation function energy value of each noisy component in the component set is obtained. Based on the autocorrelation function energy value, the dominant boundary component in the component set can be determined. Since the adjacent higher-order components of the boundary component are mainly noise, the adjacent higher-order components of the boundary component are denoised by wavelet thresholding to obtain the denoised components corresponding to the adjacent higher-order components. Finally, the denoised components and the boundary components are reconstructed to obtain the denoised time-domain difference frequency signal. By combining adaptive mode decomposition and wavelet thresholding to achieve noise filtering of the difference frequency signal, the signal-to-noise ratio of the difference frequency signal is improved. Then, fast Fourier transform processing is performed on the denoised difference frequency signal to ensure effective extraction of the difference frequency signal and improve the success rate of difference frequency extraction. By obtaining the autocorrelation function energy values ​​of each noisy component and forming an autocorrelation function energy curve, the noisy component dominated by the useful signal in the initial difference frequency signal can be quickly and accurately obtained. Through signal reconstruction processing of the difference frequency signal in different frequency bands, the signal reconstruction methods can be enriched, the accuracy of the reconstructed difference frequency signal can be improved, and thus the success rate of difference frequency extraction can be further enhanced.

[0091] See Figure 5 This provides a schematic diagram of the energy collection generation process for embodiments of this application. For example... Figure 5 As shown, the energy collection generation process is as follows: Figure 2 The execution process of step S202 in the illustrated embodiment specifically includes:

[0092] S301, obtain any two component values ​​of the target noisy component in the component set, and calculate the autocorrelation function of the target noisy component based on the component values;

[0093] S302, calculate the autocorrelation function energy value of the target noisy component based on the autocorrelation function, and add the autocorrelation function energy value of the target noisy component to the energy set of the initial difference frequency signal;

[0094] Specifically, the signal noise filtering device can obtain the autocorrelation function corresponding to each noisy component in the component set, and generate the energy set of the initial difference frequency signal based on the autocorrelation function. The energy set includes the energy value of the autocorrelation function corresponding to each noisy component in the component set. The autocorrelation function can be an unbiased autocorrelation function, which reflects the correlation between the values ​​of the signal represented by the noisy component at any two different times. Optionally, the signal noise filtering device can obtain any two component values ​​of the target noisy component in the component set. The target noisy component is any noisy component in the component set, and the component values ​​are the component values ​​corresponding to any two times of the target noisy component. The signal noise filtering device can calculate the autocorrelation function of the target noisy component based on the component values. The autocorrelation function can be expressed by the following formula:

[0095] Ri ( t 1, t 2)= E [ ci ( t 1) ci ( t 2)]

[0096] Where c represents any noisy component in the component set, i.e., the target noisy component, and t1 and t2 represent two arbitrary times in the target noisy component.

[0097] The signal-noise filtering device can calculate the autocorrelation function energy value of the target noisy component based on the autocorrelation function. The autocorrelation function energy value can be calculated using the following formula:

[0098]

[0099] Where i represents the i-th noisy component in the component set.

[0100] The signal noise filtering device can add the autocorrelation function energy value of the target noisy component to the energy set of the initial difference frequency signal. Similarly, for the remaining components in the component set, the corresponding autocorrelation function energy value can be obtained according to the above calculation process of the target noisy component and added to the energy set. The energy set may include the autocorrelation function energy value corresponding to each noisy component in the component set.

[0101] In this embodiment, by performing component decomposition on the initial difference frequency signal of the lidar, the autocorrelation function energy values ​​of each noisy component in the component set are obtained. Based on the autocorrelation function energy values, the dominant boundary component in the component set can be determined. Since the adjacent higher-order components of the boundary component are mainly noise, wavelet thresholding is used to denoise the adjacent higher-order components of the boundary component, resulting in denoised components corresponding to the adjacent higher-order components. Finally, signal reconstruction processing is performed on the denoised components and the boundary components to obtain the denoised time-domain difference frequency signal. By combining adaptive mode decomposition and wavelet thresholding to achieve the noise filtering process of the difference frequency signal, the signal-to-noise ratio of the difference frequency signal is improved. Furthermore, fast Fourier transform processing is performed on the denoised difference frequency signal to ensure effective extraction of the difference frequency signal and improve the success rate of difference frequency extraction.

[0102] See Figure 6 This provides a flowchart illustrating the process of determining the boundary components in an embodiment of this application. Figure 6 As shown, the process for determining the boundary component is as follows: Figure 2 The execution process of step S203 in the illustrated embodiment specifically includes:

[0103] S401, Generate an autocorrelation function energy curve based on the energy values ​​of the respective correlation functions in the energy set;

[0104] S402, obtain the maximum autocorrelation function energy value in the autocorrelation function energy curve, and determine the noisy component corresponding to the maximum autocorrelation function energy value as the boundary component;

[0105] Specifically, the signal noise filtering device can obtain the maximum autocorrelation function energy value in the energy set and determine the noisy component corresponding to the maximum autocorrelation function energy value as the boundary component. Optionally, the signal noise filtering device can generate an autocorrelation function energy curve based on the respective autocorrelation function energy values ​​in the energy set. The signal noise filtering device can obtain the maximum autocorrelation function energy value in the autocorrelation function energy curve and determine the noisy component corresponding to the maximum autocorrelation function energy value as the boundary component, that is, quickly and accurately obtain the noisy component dominated by the useful signal in the initial difference frequency signal. For example: when the maximum value of the autocorrelation function energy in the autocorrelation function energy curve is located on the first-order noisy component, that is, the first-order noisy component in the component set is the noisy component dominated by the useful signal in the initial difference frequency signal, the first-order noisy component is determined as the boundary component; when the maximum value of the autocorrelation function energy in the autocorrelation function energy curve is located on the k-th noisy component, that is, the k-th noisy component in the component set is the noisy component dominated by the useful signal in the initial difference frequency signal, the k-th noisy component is determined as the boundary component, where k is a positive integer greater than 1.

[0106] In this embodiment, by performing component decomposition on the initial difference frequency signal of the lidar, the autocorrelation function energy values ​​of each noisy component in the component set are obtained. Based on the autocorrelation function energy values, the dominant boundary component in the component set can be determined. Since the adjacent higher-order components of the boundary component are mainly noise, wavelet thresholding is used to denoise the adjacent higher-order components of the boundary component, obtaining the denoised components corresponding to the adjacent higher-order components. Finally, the denoised components and the boundary components are subjected to signal reconstruction processing to obtain the denoised time-domain difference frequency signal. By combining adaptive mode decomposition and wavelet thresholding to achieve the noise filtering process of the difference frequency signal, the signal-to-noise ratio of the difference frequency signal is improved. Furthermore, the fast Fourier transform processing of the denoised difference frequency signal ensures effective extraction of the difference frequency signal and improves the success rate of difference frequency extraction. By obtaining the autocorrelation function energy values ​​of each noisy component and forming an autocorrelation function energy curve, the noisy component dominated by the useful signal in the initial difference frequency signal can be quickly and accurately obtained.

[0107] In the embodiments of this application, please refer to Figure 7 and Figure 8 The figure shows the autocorrelation function energy curves under two different signal-to-noise ratios. The initial difference frequency signal is decomposed into eight noisy components. Since the noisy components are ranked according to the magnitude of their frequency fluctuations, the first to eighth noisy components are arranged from high to low frequency fluctuation. Figure 7As shown, when the signal-to-noise ratio (SNR) is -12 dBm, the maximum value of the autocorrelation function energy in the autocorrelation function energy curve is located on the first-order noisy component. That is, the first-order noisy component (also called the "first noisy component", and so on) in the component set is the noisy component dominated by the useful signal in the initial difference frequency signal. Therefore, the first-order noisy component is determined as the boundary component; for example... Figure 8 As shown, when the SNR is -5 dBm, the maximum value of the autocorrelation function energy in the autocorrelation function energy curve is located on the second-order noisy component. That is, the second-order noisy component in the component set is the noisy component dominated by the useful signal in the initial difference frequency signal. Therefore, the second-order noisy component is determined as the boundary component. Similarly, when the maximum value of the autocorrelation function energy in the autocorrelation function energy curve is located on the k-th order noisy component, that is, the k-th order noisy component in the component set is the noisy component dominated by the useful signal in the initial difference frequency signal, the k-th order noisy component is determined as the boundary component. By obtaining the autocorrelation function energy value of each noisy component and forming the autocorrelation function energy curve, the noisy component dominated by the useful signal in the initial difference frequency signal can be obtained quickly and accurately.

[0108] based on Figure 1 The system architecture will be discussed below in conjunction with the appendix. Figure 9 -Appendix Figure 11 This paper provides a detailed description of the signal and noise filtering device provided in the embodiments of this application. It should be noted that the appendix... Figure 9 -Appendix Figure 11 The signal and noise filtering device in the present application is used to perform the present application. Figures 2-8 The methods shown in the embodiments are for illustrative purposes only, illustrating the parts relevant to the embodiments of this application. For specific technical details not disclosed, please refer to this application. Figures 2-8 The example shown.

[0109] See Figure 9 This is a schematic diagram of a signal noise filtering device provided in an embodiment of this application. Figure 9 As shown, the signal noise filtering device 1 in this embodiment may include: a component set acquisition unit 11, a boundary component acquisition unit 12, a noise reduction component acquisition unit 13, and a signal reconstruction unit 14.

[0110] The component set acquisition unit 11 is used to perform ensemble empirical mode decomposition on the initial difference frequency signal generated by the lidar to obtain the component set corresponding to the initial difference frequency signal.

[0111] The boundary component acquisition unit 12 is used to acquire the autocorrelation function energy value corresponding to each noisy component in the component set, and to acquire the boundary component corresponding to the largest autocorrelation function energy value among the noisy components.

[0112] The denoising component acquisition unit 13 is used to perform wavelet threshold denoising processing on the adjacent higher-order noisy components of the boundary component to obtain the denoised components corresponding to the adjacent higher-order noisy components.

[0113] The adjacent higher-order noisy component is the noisy component in the component set that is adjacent to the boundary component and has a frequency fluctuation range greater than that of the boundary component.

[0114] The signal reconstruction unit 14 is used to perform signal reconstruction processing based on the frequency band region of the initial difference frequency signal in the spectrum, and based on the denoised component and the boundary component, to obtain the denoised time-domain difference frequency signal.

[0115] In this embodiment, by performing component decomposition on the initial difference frequency signal of the lidar, the autocorrelation function energy values ​​of each noisy component in the component set are obtained. Based on the autocorrelation function energy values, the dominant boundary component in the component set can be determined. Since the adjacent higher-order components of the boundary component are mainly noise, wavelet thresholding is used to denoise the adjacent higher-order components of the boundary component, resulting in denoised components corresponding to the adjacent higher-order components. Finally, signal reconstruction processing is performed on the denoised components and the boundary components to obtain the denoised time-domain difference frequency signal. By combining adaptive mode decomposition and wavelet thresholding to achieve the noise filtering process of the difference frequency signal, the signal-to-noise ratio of the difference frequency signal is improved. Furthermore, fast Fourier transform processing is performed on the denoised difference frequency signal to ensure effective extraction of the difference frequency signal and improve the success rate of difference frequency extraction.

[0116] See Figure 10 This is a schematic diagram of a signal noise filtering device provided in an embodiment of this application. Figure 10 As shown, the signal noise filtering device 1 in this application embodiment may include: a component set acquisition unit 11, a boundary component acquisition unit 12, a noise reduction component acquisition unit 13, a signal reconstruction unit 14, and a difference frequency acquisition unit 15.

[0117] The component set acquisition unit 11 is used to perform ensemble empirical mode decomposition on the initial difference frequency signal generated by the lidar to obtain the component set corresponding to the initial difference frequency signal.

[0118] The boundary component acquisition unit 12 is used to acquire the autocorrelation function energy value corresponding to each noisy component in the component set, and to acquire the boundary component corresponding to the largest autocorrelation function energy value among the noisy components.

[0119] For details, please refer to the following: Figure 11 The diagram below provides a structural schematic of the boundary component acquisition unit for embodiments of this application. Figure 11 As shown, the boundary component acquisition unit 12 may include:

[0120] Energy combination acquisition subunit 121 is used to acquire the autocorrelation function corresponding to each noisy component in the component set, and generate the energy set of the initial difference frequency signal based on the autocorrelation function;

[0121] In a specific implementation, the energy set includes the autocorrelation function energy value corresponding to each noisy component in the component set. The energy combination acquisition subunit 121 is specifically used to acquire any two component values ​​of the target noisy component in the component set, calculate the autocorrelation function of the target noisy component based on the component values, wherein the target noisy component is any noisy component in the component set, and the component values ​​are the component values ​​corresponding to any two times in the target noisy component; calculate the autocorrelation function energy value of the target noisy component based on the autocorrelation function, and add the autocorrelation function energy value of the target noisy component to the energy set of the initial difference frequency signal.

[0122] Boundary component determination subunit 122 is used to obtain the maximum autocorrelation function energy value in the energy set and determine the noisy component corresponding to the maximum autocorrelation function energy value as the boundary component.

[0123] In a specific implementation, the boundary component determination subunit 122 is specifically used to generate an autocorrelation function energy curve based on the respective correlation function energy values ​​in the energy set; obtain the largest autocorrelation function energy value in the autocorrelation function energy curve, and determine the noisy component corresponding to the largest autocorrelation function energy value as the boundary component.

[0124] The denoising component acquisition unit 13 is used to perform wavelet threshold denoising processing on the adjacent higher-order noisy components of the boundary component to obtain the denoised components corresponding to the adjacent higher-order noisy components.

[0125] In a specific implementation, the adjacent higher-order noisy components are the noisy components in the component set that are adjacent to the boundary component and have a frequency fluctuation range greater than that of the boundary component. The denoising component acquisition unit 13 is specifically used to perform wavelet threshold denoising on the boundary component when the boundary component is a first-order noisy component in the component set, to obtain a first denoised component corresponding to the boundary component; and to perform wavelet threshold denoising on the adjacent higher-order noisy components of the boundary component when the boundary component is a non-first-order noisy component in the component set, to obtain a second denoised component corresponding to the adjacent higher-order noisy component.

[0126] The signal reconstruction unit 14 is used to perform signal reconstruction processing based on the frequency band region of the initial difference frequency signal in the spectrum, and based on the denoised component and the boundary component, to obtain the denoised time-domain difference frequency signal.

[0127] In a specific implementation, when the boundary component is the first-order noisy component in the component set, the signal reconstruction unit 14 is specifically used to perform signal reconstruction processing on the first denoised component and the second-order noisy component in the component set when the initial difference frequency signal is in the first frequency band region in the spectrum, to obtain a denoised time-domain difference frequency signal; when the initial difference frequency signal is in the second frequency band region in the spectrum, to perform signal reconstruction processing on the first denoised component, the second-order noisy component and the third-order noisy component in the component set, to obtain a denoised time-domain difference frequency signal; when the initial difference frequency signal is in the third frequency band region in the spectrum, to perform signal reconstruction processing on the first denoised component and the remaining noisy component, to obtain a denoised time-domain difference frequency signal, wherein the remaining noisy component is the other noisy component in the component set except for the first-order noisy component.

[0128] When the boundary component is a non-first-order noisy component in the component set, the signal reconstruction unit 14 is specifically used to perform signal reconstruction processing on the second denoised component and the boundary component when the initial difference frequency signal is in the first frequency band region in the spectrum, to obtain a denoised time-domain difference frequency signal; when the initial difference frequency signal is in the second frequency band region in the spectrum, to perform signal reconstruction processing on the second denoised component, the boundary component, and the adjacent low-order noisy components of the boundary component, to obtain a denoised time-domain difference frequency signal, wherein the adjacent low-order noisy components are the noisy components in the component set that are adjacent to the boundary component and have a frequency fluctuation range smaller than the boundary component; when the initial difference frequency signal is in the third frequency band region in the spectrum, to perform signal reconstruction processing on the second denoised component, the boundary component, and the remaining low-order noisy components of the boundary component, to obtain a denoised time-domain difference frequency signal, wherein the remaining low-order noisy components are all noisy components in the component set whose frequency fluctuation range is smaller than the boundary component;

[0129] Wherein, the maximum frequency value of the second frequency band region is less than the minimum frequency value of the first frequency band region, and the minimum frequency value of the second frequency band region is greater than the maximum frequency value of the third frequency band region.

[0130] The difference frequency acquisition unit 15 is used to perform fast Fourier transform processing on the time-domain difference frequency signal to obtain the frequency-domain difference frequency signal, and to obtain the difference frequency value corresponding to the maximum amplitude value in the frequency-domain difference frequency signal.

[0131] In this embodiment, by performing component decomposition on the initial difference frequency signal of the lidar, the autocorrelation function energy value of each noisy component in the component set is obtained. Based on the autocorrelation function energy value, the dominant boundary component in the component set can be determined. Since the adjacent higher-order components of the boundary component are mainly noise, the adjacent higher-order components of the boundary component are denoised by wavelet thresholding to obtain the denoised components corresponding to the adjacent higher-order components. Finally, the denoised components and the boundary components are reconstructed to obtain the denoised time-domain difference frequency signal. By combining adaptive mode decomposition and wavelet thresholding to achieve noise filtering of the difference frequency signal, the signal-to-noise ratio of the difference frequency signal is improved. Then, fast Fourier transform processing is performed on the denoised difference frequency signal to ensure effective extraction of the difference frequency signal and improve the success rate of difference frequency extraction. By obtaining the autocorrelation function energy values ​​of each noisy component and forming an autocorrelation function energy curve, the noisy component dominated by the useful signal in the initial difference frequency signal can be quickly and accurately obtained. Through signal reconstruction processing of the difference frequency signal in different frequency bands, the signal reconstruction methods can be enriched, the accuracy of the reconstructed difference frequency signal can be improved, and thus the success rate of difference frequency extraction can be further enhanced.

[0132] This application also provides a computer storage medium that can store multiple program instructions, which are adapted to be loaded and executed by a processor as described above. Figures 2-8 The method steps of the illustrated embodiment can be found in the following documentation for detailed execution. Figures 2-8 The specific details of the illustrated embodiments will not be elaborated here.

[0133] See Figure 12 The diagram below provides a structural schematic of a lidar according to an embodiment of this application. Figure 12 As shown, the lidar 1000 may include: at least one processor 1001, such as a CPU, at least one network interface 1004, an input / output interface 1003, a memory 1005, and at least one communication bus 1002. The communication bus 1002 is used to enable communication between these components. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1005 may be high-speed RAM or non-volatile memory, such as at least one disk storage device. Optionally, the memory 1005 may also be at least one storage device located remotely from the aforementioned processor 1001. Figure 12 As shown, the memory 1005, which serves as a computer storage medium, may include an operating system, a network communication module, an input / output interface module, and a noise filtering application.

[0134] exist Figure 12 In the lidar 1000 shown, the input / output interface 1003 is mainly used to provide an input interface for users and access devices to acquire data input by users and access devices.

[0135] In one embodiment, processor 1001 can be used to invoke a noise filtering application stored in memory 1005 and specifically perform the following operations:

[0136] The initial difference frequency signal generated by the lidar is subjected to ensemble empirical mode decomposition to obtain the component set corresponding to the initial difference frequency signal;

[0137] Obtain the autocorrelation function energy value corresponding to each noisy component in the component set, and obtain the boundary component corresponding to the largest autocorrelation function energy value among the noisy components;

[0138] Wavelet threshold denoising is performed on the adjacent higher-order noisy components of the boundary component to obtain the denoised components corresponding to the adjacent higher-order noisy components. The adjacent higher-order noisy components are the noisy components in the component set that are adjacent to the boundary component and have a frequency fluctuation range greater than that of the boundary component.

[0139] Based on the frequency band region of the initial difference frequency signal in the spectrum, and based on the denoised component and the boundary component, signal reconstruction processing is performed to obtain the denoised time-domain difference frequency signal.

[0140] Optionally, when the processor 1001 performs the operation of obtaining the autocorrelation function energy value corresponding to each noisy component in the component set, and obtaining the boundary component corresponding to the largest autocorrelation function energy value among the noisy components, it specifically performs the following operations:

[0141] Obtain the autocorrelation function corresponding to each noisy component in the component set, and generate the energy set of the initial difference frequency signal based on the autocorrelation function. The energy set includes the energy value of the autocorrelation function corresponding to each noisy component in the component set.

[0142] The maximum autocorrelation function energy value is obtained from the energy set, and the noisy component corresponding to the maximum autocorrelation function energy value is determined as the boundary component.

[0143] Optionally, when the processor 1001 executes the process of obtaining the autocorrelation function corresponding to each noisy component in the component set, and obtaining the energy value of the autocorrelation function corresponding to each noisy component based on the autocorrelation function, it specifically performs the following operations:

[0144] Obtain any two component values ​​of the target noisy component in the component set, and calculate the autocorrelation function of the target noisy component based on the component values. The target noisy component is any noisy component in the component set, and the component values ​​are the component values ​​corresponding to any two times in the target noisy component.

[0145] The autocorrelation function energy value of the target noisy component is calculated based on the autocorrelation function, and the autocorrelation function energy value of the target noisy component is added to the energy set of the initial difference frequency signal.

[0146] Optionally, when the processor 1001 performs the operation of obtaining the maximum autocorrelation function energy value in the energy set and determining the noisy component corresponding to the maximum autocorrelation function energy value as the boundary component, it specifically performs the following operations:

[0147] Generate autocorrelation function energy curves based on the energy values ​​of the respective correlation functions in the energy set;

[0148] The maximum autocorrelation function energy value is obtained from the autocorrelation function energy curve, and the noisy component corresponding to the maximum autocorrelation function energy value is determined as the boundary component.

[0149] Optionally, when the processor 1001 performs wavelet threshold denoising on the adjacent higher-order noisy components of the boundary component to obtain the denoised components corresponding to the adjacent higher-order noisy components, it specifically performs the following operations:

[0150] When the boundary component is the first-order noisy component in the component set, wavelet threshold denoising is performed on the boundary component to obtain the first denoised component corresponding to the boundary component.

[0151] When the boundary component is a non-first-order noisy component in the component set, wavelet threshold denoising is performed on the adjacent higher-order noisy components of the boundary component to obtain the second denoised component corresponding to the adjacent higher-order noisy component.

[0152] Optionally, when the boundary component is a first-order noisy component in the component set, when the processor 1001 performs signal reconstruction processing based on the frequency band region of the initial difference frequency signal in the spectrum and based on the denoised component and the boundary component to obtain the denoised time-domain difference frequency signal, it specifically performs the following operations:

[0153] When the initial difference frequency signal is in the first frequency band region in the spectrum, signal reconstruction processing is performed on the first denoised component and the second-order noisy component in the component set to obtain the denoised time-domain difference frequency signal.

[0154] When the initial difference frequency signal is in the second frequency band region in the spectrum, signal reconstruction processing is performed on the first denoised component, the second-order noisy component and the third-order noisy component in the component set to obtain the denoised time-domain difference frequency signal.

[0155] When the initial difference frequency signal is in the third frequency band region in the spectrum, the first denoised component and the remaining noisy component are subjected to signal reconstruction processing to obtain the denoised time-domain difference frequency signal. The remaining noisy component is the other noisy component in the component set except for the first-order noisy component.

[0156] Wherein, the maximum frequency value of the second frequency band region is less than the minimum frequency value of the first frequency band region, and the minimum frequency value of the second frequency band region is greater than the maximum frequency value of the third frequency band region.

[0157] Optionally, when the boundary component is a non-first-order noisy component in the component set, when the processor 1001 performs signal reconstruction processing based on the frequency band region of the initial difference frequency signal in the spectrum and based on the denoised component and the boundary component to obtain the denoised time-domain difference frequency signal, it specifically performs the following operations:

[0158] When the initial difference frequency signal is in the first frequency band region in the spectrum, the second denoised component and the boundary component are subjected to signal reconstruction processing to obtain the denoised time-domain difference frequency signal.

[0159] When the initial difference frequency signal is in the second frequency band region in the spectrum, the second denoised component, the boundary component, and the adjacent low-order noisy components of the boundary component are subjected to signal reconstruction processing to obtain the denoised time-domain difference frequency signal. The adjacent low-order noisy components are the noisy components in the component set that are adjacent to the boundary component and have a frequency fluctuation range smaller than that of the boundary component.

[0160] When the initial difference frequency signal is in the third frequency band region in the spectrum, the second denoised component, the boundary component, and the remaining low-order noisy component of the boundary component are subjected to signal reconstruction processing to obtain the denoised time-domain difference frequency signal. The remaining low-order noisy component is all noisy components in the component set whose frequency fluctuation range is smaller than that of the boundary component.

[0161] Wherein, the maximum frequency value of the second frequency band region is less than the minimum frequency value of the first frequency band region, and the minimum frequency value of the second frequency band region is greater than the maximum frequency value of the third frequency band region.

[0162] Optionally, the processor 1001 also performs the following operations:

[0163] The time-domain difference frequency signal is processed by a fast Fourier transform to obtain a frequency-domain difference frequency signal, and the difference frequency value corresponding to the maximum amplitude value is obtained from the frequency-domain difference frequency signal.

[0164] In this embodiment, by performing component decomposition on the initial difference frequency signal of the lidar, the autocorrelation function energy value of each noisy component in the component set is obtained. Based on the autocorrelation function energy value, the dominant boundary component in the component set can be determined. Since the adjacent higher-order components of the boundary component are mainly noise, the adjacent higher-order components of the boundary component are denoised by wavelet thresholding to obtain the denoised components corresponding to the adjacent higher-order components. Finally, the denoised components and the boundary components are reconstructed to obtain the denoised time-domain difference frequency signal. By combining adaptive mode decomposition and wavelet thresholding to achieve noise filtering of the difference frequency signal, the signal-to-noise ratio of the difference frequency signal is improved. Then, fast Fourier transform processing is performed on the denoised difference frequency signal to ensure effective extraction of the difference frequency signal and improve the success rate of difference frequency extraction. By obtaining the autocorrelation function energy values ​​of each noisy component and forming an autocorrelation function energy curve, the noisy component dominated by the useful signal in the initial difference frequency signal can be quickly and accurately obtained. Through signal reconstruction processing of the difference frequency signal in different frequency bands, the signal reconstruction methods can be enriched, the accuracy of the reconstructed difference frequency signal can be improved, and thus the success rate of difference frequency extraction can be further enhanced.

[0165] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0166] The above-disclosed embodiments are merely preferred embodiments of this application and should not be construed as limiting the scope of this application. Therefore, any equivalent variations made in accordance with the claims of this application shall still fall within the scope of this application.

Claims

1. A method for signal noise filtering, characterized in that, include: The initial difference frequency signal generated by the lidar is subjected to ensemble empirical mode decomposition to obtain the component set corresponding to the initial difference frequency signal; Obtain the autocorrelation function energy value corresponding to each noisy component in the component set, and obtain the boundary component corresponding to the largest autocorrelation function energy value among the noisy components; Wavelet threshold denoising is performed on the adjacent higher-order noisy components of the boundary component to obtain the denoised components corresponding to the adjacent higher-order noisy components. The adjacent higher-order noisy components are the noisy components in the component set that are adjacent to the boundary component and have a frequency fluctuation range greater than that of the boundary component. Based on the frequency band region of the initial difference frequency signal in the spectrum, and based on the denoised component and the boundary component, signal reconstruction processing is performed to obtain the denoised time-domain difference frequency signal. The step of performing wavelet threshold denoising on the adjacent higher-order noisy components of the boundary component to obtain the denoised components corresponding to the adjacent higher-order noisy components includes: When the boundary component is a non-first-order noisy component in the component set, wavelet threshold denoising is performed on the adjacent higher-order noisy components of the boundary component to obtain the second denoised component corresponding to the adjacent higher-order noisy component. The component set includes first-order noisy components to k-order noisy components, and the non-first-order noisy component is any one of the second-order noisy components to k-order noisy components, where k is a positive integer greater than 1. When the boundary component is a non-first-order noisy component in the component set, the signal reconstruction process based on the frequency band region of the initial difference frequency signal in the spectrum, and based on the denoised component and the boundary component, to obtain the denoised time-domain difference frequency signal includes: When the initial difference frequency signal is in the first frequency band region in the spectrum, the second denoised component and the boundary component are subjected to signal reconstruction processing to obtain the denoised time-domain difference frequency signal. When the initial difference frequency signal is in the second frequency band region in the spectrum, the second denoised component, the boundary component, and the adjacent low-order noisy components of the boundary component are subjected to signal reconstruction processing to obtain the denoised time-domain difference frequency signal. The adjacent low-order noisy components are the noisy components in the component set that are adjacent to the boundary component and have a frequency fluctuation range smaller than that of the boundary component. When the initial difference frequency signal is in the third frequency band region in the spectrum, the second denoised component, the boundary component, and the remaining low-order noisy component of the boundary component are subjected to signal reconstruction processing to obtain the denoised time-domain difference frequency signal. The remaining low-order noisy component is all noisy components in the component set whose frequency fluctuation range is smaller than that of the boundary component. Wherein, the maximum frequency value of the second frequency band region is less than the minimum frequency value of the first frequency band region, and the minimum frequency value of the second frequency band region is greater than the maximum frequency value of the third frequency band region.

2. The method according to claim 1, characterized in that, The step of obtaining the autocorrelation function energy value corresponding to each noisy component in the component set, and obtaining the boundary component corresponding to the largest autocorrelation function energy value among the noisy components, includes: Obtain the autocorrelation function corresponding to each noisy component in the component set, and generate the energy set of the initial difference frequency signal based on the autocorrelation function. The energy set includes the energy value of the autocorrelation function corresponding to each noisy component in the component set. The maximum autocorrelation function energy value is obtained from the energy set, and the noisy component corresponding to the maximum autocorrelation function energy value is determined as the boundary component.

3. The method according to claim 2, characterized in that, The step of obtaining the autocorrelation function corresponding to each noisy component in the component set, and obtaining the autocorrelation function energy value corresponding to each noisy component based on the autocorrelation function, includes: Obtain any two component values ​​of the target noisy component in the component set, and calculate the autocorrelation function of the target noisy component based on the component values. The target noisy component is any noisy component in the component set, and the component values ​​are the component values ​​corresponding to any two times in the target noisy component. The autocorrelation function energy value of the target noisy component is calculated based on the autocorrelation function, and the autocorrelation function energy value of the target noisy component is added to the energy set of the initial difference frequency signal.

4. The method according to claim 2, characterized in that, The step of obtaining the maximum autocorrelation function energy value in the energy set and determining the noisy component corresponding to the maximum autocorrelation function energy value as the boundary component includes: Generate autocorrelation function energy curves based on the energy values ​​of the respective correlation functions in the energy set; The maximum autocorrelation function energy value is obtained from the autocorrelation function energy curve, and the noisy component corresponding to the maximum autocorrelation function energy value is determined as the boundary component.

5. The method according to claim 1, wherein The step of performing wavelet threshold denoising on the adjacent higher-order noisy components of the boundary component to obtain the denoised components corresponding to the adjacent higher-order noisy components includes: When the boundary component is the first noisy component in the component set, wavelet threshold denoising is performed on the boundary component to obtain the first denoised component corresponding to the boundary component.

6. The method according to claim 5, characterized in that, When the boundary component is the first-order noisy component in the component set, the step of performing signal reconstruction processing based on the frequency band region of the initial difference frequency signal in the spectrum, and based on the denoised component and the boundary component, to obtain the denoised time-domain difference frequency signal includes: When the initial difference frequency signal is in the first frequency band region in the spectrum, signal reconstruction processing is performed on the first denoised component and the second-order noisy component in the component set to obtain the denoised time-domain difference frequency signal. When the initial difference frequency signal is in the second frequency band region in the spectrum, signal reconstruction processing is performed on the first denoised component, the second-order noisy component and the third-order noisy component in the component set to obtain the denoised time-domain difference frequency signal. When the initial difference frequency signal is in the third frequency band region in the spectrum, the first denoised component and the remaining noisy component are subjected to signal reconstruction processing to obtain the denoised time-domain difference frequency signal. The remaining noisy component is the other noisy component in the component set except for the first-order noisy component. Wherein, the maximum frequency value of the second frequency band region is less than the minimum frequency value of the first frequency band region, and the minimum frequency value of the second frequency band region is greater than the maximum frequency value of the third frequency band region.

7. The method according to claim 1, characterized in that, Also includes: The time-domain difference frequency signal is processed by a fast Fourier transform to obtain a frequency-domain difference frequency signal, and the difference frequency value corresponding to the maximum amplitude value is obtained from the frequency-domain difference frequency signal.

8. A signal noise filtering device, characterized in that, include: The component set acquisition unit is used to perform ensemble empirical mode decomposition on the initial difference frequency signal generated by the lidar to obtain the component set corresponding to the initial difference frequency signal. The boundary component acquisition unit is used to acquire the autocorrelation function energy value corresponding to each noisy component in the component set, and to acquire the boundary component corresponding to the largest autocorrelation function energy value among the noisy components. The denoising component acquisition unit is used to perform wavelet threshold denoising processing on the adjacent higher-order noisy components of the boundary component to obtain the denoised components corresponding to the adjacent higher-order noisy components. The adjacent higher-order noisy components are noisy components in the component set that are adjacent to the boundary component and have a frequency fluctuation range greater than that of the boundary component. The signal reconstruction unit is used to perform signal reconstruction processing based on the frequency band region of the initial difference frequency signal in the spectrum, and based on the denoised component and the boundary component, to obtain the denoised time-domain difference frequency signal. The step of performing wavelet threshold denoising on the adjacent higher-order noisy components of the boundary component to obtain the denoised components corresponding to the adjacent higher-order noisy components includes: When the boundary component is a non-first-order noisy component in the component set, wavelet threshold denoising is performed on the adjacent higher-order noisy components of the boundary component to obtain the second denoised component corresponding to the adjacent higher-order noisy component. The component set includes first-order noisy components to k-order noisy components, and the non-first-order noisy component is any one of the second-order noisy components to k-order noisy components, where k is a positive integer greater than 1. When the boundary component is a non-first-order noisy component in the component set, the signal reconstruction process based on the frequency band region of the initial difference frequency signal in the spectrum, and based on the denoised component and the boundary component, to obtain the denoised time-domain difference frequency signal includes: When the initial difference frequency signal is in the first frequency band region in the spectrum, the second denoised component and the boundary component are subjected to signal reconstruction processing to obtain the denoised time-domain difference frequency signal. When the initial difference frequency signal is in the second frequency band region in the spectrum, the second denoised component, the boundary component, and the adjacent low-order noisy components of the boundary component are subjected to signal reconstruction processing to obtain the denoised time-domain difference frequency signal. The adjacent low-order noisy components are the noisy components in the component set that are adjacent to the boundary component and have a frequency fluctuation range smaller than that of the boundary component. When the initial difference frequency signal is in the third frequency band region in the spectrum, the second denoised component, the boundary component, and the remaining low-order noisy component of the boundary component are subjected to signal reconstruction processing to obtain the denoised time-domain difference frequency signal. The remaining low-order noisy component is all noisy components in the component set whose frequency fluctuation range is smaller than that of the boundary component. Wherein, the maximum frequency value of the second frequency band region is less than the minimum frequency value of the first frequency band region, and the minimum frequency value of the second frequency band region is greater than the maximum frequency value of the third frequency band region.

9. A lidar, characterized in that, Includes processor, memory, and input / output interfaces; The processor is connected to the memory and the input / output interface respectively, wherein the input / output interface is used for page interaction, the memory is used to store program code, and the processor is used to call the program code to execute the method as described in any one of claims 1-7.

Images