A method and device for matching, fusing and inverting lidar signals

Abstract

The application relates to the technical field of laser radar remote sensing detection, in particular to a laser radar signal matching fusion and inversion method and device. First, the application acquires first echo signals and second echo signals after preprocessing, which correspond to wide field of view near-ground detection and narrow field of view high-altitude detection respectively. Then, in a preset height range, the best matching area is autonomously determined by step-by-step search, comprehensively considering the correlation coefficient and signal strength difference. Next, according to the upper and lower boundaries of the best matching area, the whole area is divided into a high-altitude section, a fusion section and a near-ground section, a linear mapping relationship is established by using orthogonal regression fitting, and segmented fusion is carried out. Finally, based on the fusion signal, the concentration profile of the characteristic pollutant in the whole area detection height range is obtained. The application realizes autonomous matching and seamless fusion of double field of view signals, avoids data mutation in the transition zone, and significantly improves the pollutant detection precision and the whole area coverage capability of the differential absorption laser radar from the near-ground to the high altitude.

Description

Technical Field

[0001] This invention relates to the field of lidar remote sensing technology, and in particular to a lidar signal matching, fusion and inversion method and apparatus. Background Technology

[0002] Differential absorption lidar is a high-precision active optical remote sensing technology. It uses two laser beams with wavelengths located at the strong absorption peak and weak absorption valley of the target gas and compares the differences in their echo signals to accurately invert the concentration distribution of specific gases (such as ozone and sulfur dioxide) in the atmosphere. It is widely used in meteorological monitoring, air pollution monitoring and other fields.

[0003] In practical detection, the distribution of characteristic absorbing gases in the vertical direction is complex, requiring the system to simultaneously possess high-quality detection capabilities for both high-altitude and near-ground signals. However, limited by the dynamic range of the detection system, a single receiving telescope cannot handle both simultaneously. To address this, existing technologies have developed dual-field-of-view lidar systems. These systems typically employ two sets of receiving telescopes: the first telescope (e.g., a small-aperture telescope) has a wide field of view, effectively reducing the near-surface detection blind zone and acquiring high-quality short-range backscattered signals, but its effective detection altitude is limited; the second telescope (e.g., a large-aperture telescope) uses a narrower field of view design, has strong sky background noise suppression capabilities, and can capture high-quality signals from high-altitude areas, but has a larger near-ground detection blind zone.

[0004] Using either of the aforementioned signal sources alone is insufficient to completely and comprehensively invert the vertical profile of the characteristic gas. Therefore, effectively coordinating and fusing these two complementary signals to achieve high-precision inversion of the characteristic gas across the entire altitude range is a pressing technical problem in this field. Existing technologies lack an effective method for autonomously and accurately matching and fusing dual-field-of-view signals while ensuring a smooth transition region without abrupt changes. Summary of the Invention

[0005] This invention provides a method and apparatus for matching, fusing and inverting lidar signals, which solves the technical problem in the prior art that dual-field-of-view signals cannot be autonomously and accurately matched and fused, resulting in the inability to achieve continuous and high-precision inversion of atmospheric pollutant concentrations from near-ground to high altitude.

[0006] This specification provides an embodiment of a lidar signal matching, fusion, and inversion method, including:

[0007] The first echo signal and the second echo signal after preprocessing are acquired respectively. The first echo signal is the signal obtained by using a wide field-of-view receiving telescope to detect the first field-of-view detection area, and the second echo signal is the signal obtained by using a narrow field-of-view receiving telescope to detect the second field-of-view detection area.

[0008] Within a preset height range, the overlapping height interval between the first echo signal and the second echo signal with the highest correlation and the smallest signal strength difference is determined by step search, and the overlapping height interval is marked as the best matching area.

[0009] Based on the upper and lower boundaries of the optimal matching area, the overall detection altitude is divided into a high-altitude segment, a fusion segment, and a near-ground segment. Different fusion strategies are used to fuse the first echo signal and the second echo signal to obtain a fused signal.

[0010] The concentration profile of the characteristic pollutant within the entire detection height range is obtained based on the fused signal inversion.

[0011] Optionally, the first field of view detection area is a region for detecting near-ground range signals, and the second field of view detection area is a region for detecting high-altitude range signals; the preprocessing includes: performing distance correction, background noise suppression, and validity discrimination on the first echo signal and the second echo signal, and performing adaptive cascaded Savitzky-Golay filtering noise reduction processing on the signal that passes the validity discrimination.

[0012] Optionally, the signal validity determination includes:

[0013] Determine whether there is electromagnetic interference and inter-channel crosstalk in the first echo signal and the second echo signal. If so, mark the corresponding echo signal as invalid data.

[0014] Optionally, the signal validity determination further includes:

[0015] Calculate the signal-to-noise ratio (SNR) of the first echo signal and the second echo signal at each distance, and determine the signal with an SNR greater than a preset threshold as valid data.

[0016] Optionally, the signal validity determination further includes: real-time monitoring of the received signal strengths of the first echo signal and the second echo signal;

[0017] The gain parameter of the photodetector used to convert optical signals into electrical signals is dynamically adjusted or an optical attenuator is inserted based on the received signal strength, so that the received signal strength is always below a preset saturation threshold.

[0018] Optionally, the adaptive cascaded Savitzky-Golay filtering noise reduction process includes:

[0019] The window signal-to-noise ratio and gradient rate of change of the echo signal determined to be valid are calculated in real time; the gradient rate of change is obtained by calculating the mean of the absolute values ​​of the differential absorption of adjacent sampling points of the echo signal determined to be valid.

[0020] The signal characteristics of the effective echo signal are dynamically evaluated based on the window signal-to-noise ratio and gradient change rate, and the window length and polynomial order of the Savitzky-Golay filter are adaptively set according to the evaluation results.

[0021] Return to the process of performing real-time calculations of the window signal-to-noise ratio and gradient rate of change of the valid echo signal until the signal-to-noise ratio increment of the valid echo signal at the maximum detection distance is less than a preset threshold.

[0022] Optionally, the upper boundary of the optimal matching region is defined as the upper limit of the effective detection height of the second echo signal, and the lower boundary is the lowest height at which the first echo signal completely enters the telescope's field of view;

[0023] The step-by-step search determines the overlap height interval between the first and second echo signals, where the correlation is highest and the signal strength difference is smallest. This overlap height interval is then marked as the optimal matching region.

[0024] Starting from the lower boundary, the matching interval is gradually raised by a fixed length according to a preset step size until the upper boundary is reached. The correlation coefficient between the first echo signal and the second echo signal, as well as the average value of the absolute value of the signal strength difference, are calculated in each matching interval.

[0025] Select the overlap height interval with the smallest screening coefficient that is greater than the first preset threshold and is determined by the average value of the absolute value of the signal strength difference and the correlation coefficient, and mark the overlap height interval as the best matching region.

[0026] Optionally, the fusion of the first echo signal and the second echo signal using different fusion strategies includes:

[0027] For the high-altitude segment, the second echo signal is directly used as the inversion data source for that segment;

[0028] For the fusion segment, an orthogonal regression fitting method is used to establish a linear mapping relationship between the first echo signal and the second echo signal, and based on this linear mapping relationship, the first echo signal in the fusion segment is converted into data that matches the second echo signal;

[0029] For the near-ground segment, the linear mapping relationship obtained by fitting the fusion segment is directly used to map the first echo signal into the equivalent signal of the near-ground segment.

[0030] Optionally, after the signal conversion of the fusion segment is completed, a distance weighting factor is introduced to smooth the signal values ​​at the junction of the fusion segment and the high-altitude segment and the near-ground segment.

[0031] This specification also provides an embodiment of a lidar signal matching, fusion, and inversion device, comprising:

[0032] The signal acquisition module is used to acquire the pre-processed first echo signal and the second echo signal respectively. The first echo signal is the signal obtained by using a wide field-of-view receiving telescope to detect the first field-of-view detection area, and the second echo signal is the signal obtained by using a narrow field-of-view receiving telescope to detect the second field-of-view detection area.

[0033] The optimal matching region determination module is used to determine the overlapping height interval between the first echo signal and the second echo signal with the highest correlation and the smallest signal strength difference within a preset height range through step search, and to mark the overlapping height interval as the optimal matching region.

[0034] The signal fusion module is used to divide the global detection altitude into a high-altitude segment, a fusion segment, and a near-ground segment based on the upper and lower boundaries of the optimal matching area, and to fuse the first echo signal and the second echo signal using different fusion strategies to obtain a fused signal.

[0035] The inversion module is used to invert the concentration profile of the characteristic pollutant within the global detection height range based on the fused signal.

[0036] An electronic device includes a memory and a processor, the memory storing computer instructions, and the processor being configured to execute the computer instructions to perform the method described above.

[0037] A storage medium, characterized in that the storage medium stores computer instructions, the computer instructions being configured to execute the method described above at runtime.

[0038] This invention achieves seamless connection of dual-field-of-view signals by autonomously searching for the best matching region and adopting a segmented fusion strategy, avoiding data abrupt changes in the transition zone, thereby obtaining a continuous and high-precision pollutant concentration profile from near the ground to high altitude, significantly improving the all-domain detection capability of lidar. Attached Figure Description

[0039] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0040] Figure 1 A schematic diagram illustrating the principle of a lidar signal matching, fusion, and inversion method provided in the embodiments of this specification;

[0041] Figure 2A structural diagram of a lidar signal matching, fusion, and inversion device provided in the embodiments of this specification;

[0042] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this specification;

[0043] Figure 4 This is a schematic diagram of a computer-readable medium provided for embodiments of this specification. Detailed Implementation

[0044] The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0045] It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention.

[0046] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the invention or its application or use.

[0047] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0048] In all the examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0049] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0050] Reference Figure 1This specification provides a schematic diagram of the principle of a lidar signal matching, fusion, and inversion method, comprising: S101: acquiring a pre-processed first echo signal and a second echo signal, wherein the first echo signal is a signal obtained by detecting a first field-of-view detection area using a wide-field-of-view receiving telescope, and the second echo signal is a signal obtained by detecting a second field-of-view detection area using a narrow-field-of-view receiving telescope; S102: within a preset height range, determining the overlapping height interval with the highest correlation and smallest signal strength difference between the first echo signal and the second echo signal through a step-by-step search, and marking the overlapping height interval as the optimal matching area; S103: based on the upper and lower boundaries of the optimal matching area, dividing the entire detection height into a high-altitude segment, a fusion segment, and a near-ground segment, and using different fusion strategies to fuse the first echo signal and the second echo signal to obtain a fused signal; S104: inverting the concentration profile of characteristic pollutants within the entire detection height range based on the fused signal.

[0051] Specifically, acquire the preprocessed dual-field-of-view signals:

[0052] The system emits wavelengths at a fixed pulse repetition frequency located at the ozone strong absorption peak ( ) and weak absorption valley ( The wide-field telescope receives backscattered light from near the ground (0–3 km), which is converted into a first echo signal by a photomultiplier tube; the narrow-field telescope receives backscattered light from high altitudes (2 km–10 km), which is converted into a second echo signal. Both signals are then corrected for distance (based on the speed of light). and transmit / receive time difference Calculate distance The background noise is suppressed and the effectiveness is judged. Then, after noise reduction by adaptive Savitzky-Golay filtering, it is used as the first and second echo signals after preprocessing.

[0053] Search for the best matching region:

[0054] Let the effective detection lower boundary of the first echo signal be... (The lowest height at which the wide field-of-view signal fully enters the field of view, e.g., 200m), the effective upper boundary for detecting the second echo signal is... (At altitudes where the signal-to-noise ratio drops below 3 in narrow field of view, e.g., 5km). With a fixed length The overlapping height range is a sliding window, from Start with step size Slide up gradually to The correlation coefficient between the two signals is calculated at each window position. and average strength difference Select CC > 0.95 and the filter coefficient. The smallest window is taken as the best matching region, and its upper and lower boundaries are denoted as . and .

[0055] Segmented fusion strategy:

[0056] High-altitude section ( ): Directly using the second echo signal As a fusion signal.

[0057] Fusion segment ( The first echo signal was established using orthogonal regression fitting. To the second echo signal linear mapping The optimal parameters a and b are solved iteratively using all data points within the interval, and then... Substitution .

[0058] Nearby area ( ): Using the same linear mapping, Directly convert .

[0059] Finally, a distance weighting factor is introduced at the boundary for smoothing correction.

[0060] Inverted concentration profile:

[0061] Using the principle of differential absorption, the two wavelengths after fusion are calculated ( and The signal ratio of ozone concentration is used to retrieve the vertical profile of ozone concentration by combining auxiliary data such as atmospheric pressure and temperature.

[0062] This embodiment achieves a complete closed loop from signal acquisition, autonomous matching region search, segmented fusion to concentration inversion. Compared with existing technologies, it avoids the subjectivity of manually selecting the matching region, solves the abrupt change problem caused by directly splicing two signals, and makes the inversion profile continuous, smooth, and physically consistent throughout the entire range from near the ground to high altitude. The full-altitude detection blind zone is reduced to below 200 meters, and the high-altitude detection altitude is increased to above 10 kilometers.

[0063] Optionally, the first field of view detection area is a region for detecting near-ground range signals, and the second field of view detection area is a region for detecting high-altitude range signals; the preprocessing includes: performing distance correction, background noise suppression, and validity discrimination on the first echo signal and the second echo signal, and performing adaptive cascaded Savitzky-Golay filtering noise reduction processing on the signal that passes the validity discrimination.

[0064] Taking a set of actual ozone radar signals as an example:

[0065] (1) Distance correction operation:

[0066] The system records the emission timestamp of each laser pulse. and echo signal arrival timestamp The original distance was calculated based on the speed of light c = 299,792,458 m / s. The echo signal strength Resampling is performed at equal intervals with a resolution of 7.5 meters (corresponding to a 50ns sampling interval). For distance intervals with no valid echo signal (such as signal strength below 3 standard deviations of the system noise floor and continuously exceeding 100 meters), the data points in that segment are deducted and marked as blank.

[0067] (2) Background noise suppression processing:

[0068] Analyze the original signal attenuation characteristic curve and select a target distance range where the signal strength tends to stabilize with altitude (e.g., 10km–15km, where there is no strong gas absorption and aerosol backscattering is negligible). Extract the arithmetic mean of all echo signal intensities within this range as the background noise baseline. The net signal is obtained by subtracting the background noise baseline point by point from the original signal. If the net signal at a certain point If the value is negative, set it to zero.

[0069] (3) Validity judgment:

[0070] Electromagnetic interference and crosstalk judgment, signal-to-noise ratio (SNR) screening, and anti-saturation control are performed separately. Only signal points that simultaneously meet the requirements of no electromagnetic interference / crosstalk, SNR > 3, and no saturation are marked as valid signals.

[0071] (4) Adaptive cascaded Savitzky-Golay filter noise reduction processing:

[0072] Adaptive filtering is performed on the valid signal to output a preprocessed first echo signal and a second echo signal with improved signal-to-noise ratio and preserved features.

[0073] This embodiment discloses a complete four-step preprocessing process: distance correction ensures spatial positioning accuracy (error less than 1 meter); background noise suppression eliminates dark current and DC components from the sky background; validity discrimination removes various abnormal data; and adaptive filtering balances noise smoothing and detail preservation. Through the above processing, the signal-to-noise ratio of the original signal is improved by an average of more than 10dB, laying a high-fidelity data foundation for subsequent matching and fusion.

[0074] Optionally, the signal validity determination includes: determining whether there is electromagnetic interference and inter-channel crosstalk in the first echo signal and the second echo signal; if so, marking the corresponding echo signal as invalid data.

[0075] The system acquires the time-domain waveforms of the signals from two channels in real time at a sampling rate of 1 GHz during the laser emission interval (the interval without echo signals, such as at an altitude of over 100 km), and simultaneously performs a fast Fourier transform to obtain the spectrum.

[0076] (1) Electromagnetic interference judgment:

[0077] If a spike (peak value exceeding 10 times the average noise level and duration less than 10ns) appears in the time-domain waveform that is unrelated to the laser pulse repetition frequency (e.g., 100Hz), or if unexpected frequency components such as 50Hz and its harmonics (power frequency interference), or specific radio frequency frequencies appear in the spectrum, it is determined to be electromagnetic interference. Once electromagnetic interference is determined to exist, all echo signals (including both channels) within 1 millisecond before and after the time of interference occurrence are marked as invalid signals and are not involved in any subsequent processing.

[0078] (2) Inter-channel crosstalk judgment:

[0079] Calculate the signal correlation between the two channels in the echo-free region (e.g., above 100km). Under normal circumstances, the noise of the two channels is independent of each other, and the correlation coefficient should be close to 0. If the absolute value of the calculated correlation coefficient is greater than 0.8, and the time-domain waveform shows simultaneous and identical abnormal fluctuations, then inter-channel crosstalk is identified (e.g., due to ground loops or power supply coupling). In this case, the entire signal of both channels is marked as invalid, and a system self-test is triggered.

[0080] The above method clarifies the judgment criteria and result processing actions (marking as invalid signals). By monitoring the signal characteristics of the echo-free interval in real time, it is possible to effectively identify external electromagnetic interference (such as radar and wireless communication) and internal crosstalk faults, prevent false signals from contaminating the fusion results, and ensure the authenticity and reliability of the inversion data.

[0081] Optionally, the signal validity determination further includes: calculating the signal-to-noise ratio of the first echo signal and the second echo signal at each distance, and determining signals with a signal-to-noise ratio value greater than a preset determination threshold as valid data.

[0082] Specifically, for the net signal after distance correction and background noise suppression First, calculate the standard deviation of the background noise. Select the net signal values ​​of all sampling points within an echo-free signal range (e.g., 100km–120km). These values ​​should theoretically be zero, but in reality represent noise fluctuations. Calculate the root mean square (RMS) value of the signal within this range as the standard deviation of the background noise. : ,in This represents the total number of sampling points within that interval. For the first The distance between each sampling point.

[0083] Then, for each detection distance Net signal at the location Calculate the signal-to-noise ratio .in, Let R be the signal-to-noise ratio at a distance R. Distance The net signal strength after background noise subtraction. This represents the standard deviation of the background noise.

[0084] The preset judgment threshold is set to 3 (derived from the signal-to-noise ratio detection theory of radar equations). When When, determine the distance. The signal at that location is a valid signal; when If the signal is invalid, the data at that point will be discarded in subsequent processing.

[0085] For differential absorption lidar, since simultaneous processing is required... and For signals of two wavelengths, this step performs the above discrimination on the signals of two wavelengths respectively, and only the signals of two wavelengths are retained if both wavelengths are determined to be valid distance points.

[0086] By calculating the signal-to-noise ratio point by point, it is possible to quantitatively distinguish between signal-dominant and noise-dominant regions, automatically eliminating data with significant noise interference at distant points, and ensuring high accuracy of the fused signal. The threshold of 3 ensures a false alarm rate of less than 1%.

[0087] Optionally, the signal validity determination further includes: real-time monitoring of the received signal strength of the first echo signal and the second echo signal; dynamically adjusting the gain parameter of the photodetector used to convert the optical signal into an electrical signal or inserting an optical attenuator according to the received signal strength, so that the received signal strength is always lower than a preset saturation threshold.

[0088] The specific implementation is as follows: The system monitors the peak intensity of the first and second echo signals in real time. Peak intensity is defined as the maximum amplitude value (unit: mV) of a single pulse echo signal. The preset saturation threshold is 80% of the upper limit of the linear response range of the photodetector. The monitoring frequency is synchronized with the laser pulse repetition frequency (e.g., 100Hz), meaning the peak intensity is calculated once per pulse cycle.

[0089] Dynamic adjustment operation:

[0090] Adjusting the photodetector gain parameters: If the peak intensity of a certain channel exceeds 80mV, reduce the gain control voltage of the photomultiplier tube in that channel through the digital-to-analog converter, adjusting in 1% increments of the current gain each time, until the peak intensity falls below 80mV. If the peak intensity remains below 20mV, the gain can be appropriately increased to enhance weak signals.

[0091] Inserting an optical attenuator: If the peak intensity cannot be controlled below 80mV even after the gain has been reduced to the minimum, an optical attenuator with a known attenuation ratio (such as 10dB or 20dB) is inserted into the receiving optical path by driving a motor, and the current attenuation value is recorded for subsequent signal recovery.

[0092] The above methods ensure that the photodetector always operates in the linear region, avoiding clipping distortion caused by excessively strong signals, thus ensuring linear accuracy of the signal strength ratio in subsequent signal-to-noise ratio calculations and differential absorption inversion. At the same time, dynamic adjustment expands the dynamic range of the system, adapting to large-span signal variations.

[0093] Optionally, the adaptive cascaded Savitzky-Golay filter noise reduction process includes: real-time calculation of the window signal-to-noise ratio and gradient rate of change of the echo signal determined to be valid; the gradient rate of change is obtained by calculating the mean of the absolute values ​​of the differential absorption of adjacent sampling points of the echo signal determined to be valid; dynamically evaluating the signal characteristics of the valid echo signal based on the window signal-to-noise ratio and gradient rate of change, and adaptively setting the window length and polynomial order of the Savitzky-Golay filter based on the evaluation results; returning to execute the real-time calculation process of the window signal-to-noise ratio and gradient rate of change of the echo signal determined to be valid until the signal-to-noise ratio increment of the valid echo signal at the maximum detection distance is less than a preset threshold.

[0094] Let the valid echo signal sequence be... , Corresponding to equally spaced points. The adaptive filtering process is as follows:

[0095] Real-time calculation of window signal-to-noise ratio and gradient change rate: Select an initial sliding window length of... =11 points (corresponding to approximately 82.5 meters in height). Within the window, the window signal-to-noise ratio... =(Mean echo signal within the window) / (Standard deviation of echo signal within the window). Gradient rate of change This is the average of the absolute values ​​of the signal strength differences between adjacent sampling points. The number of sampling points within the window. Let be the echo signal strength at the j-th sampling point within the window.

[0096] Dynamically evaluate signal characteristics and calculate evaluation coefficients. : =( Normalized value) × ( (Normalized value). Normalization was performed using the historical maximum value method.

[0097] in accordance with Adaptive setting of SG filter parameters: when At time: the SG filter window is set to 11 points, and the polynomial order is 3; when At time: the SG filter window is set to 21 points, and the polynomial order is 2; when At time: the SG filter window is set to 31 points and the polynomial order is 1.

[0098] Iterative filtering: Performing one SG filter yields a new signal. Calculate the maximum detection range. Signal-to-noise ratio increment at the location .like Then Repeat the above steps as a new input signal; if If the iteration stops, the current signal is output.

[0099] By dynamically matching the filtering intensity with dual indicators, over-smoothing or under-smoothing is avoided. The cascaded iterative design brings the signal-to-noise ratio increment at the maximum detection distance to below 0.5, ensuring consistent filtering performance.

[0100] Optionally, when the signal feature evaluation coefficient is less than a first threshold, the filtering window is set to a first window length and the polynomial order is set to a first order; when the evaluation coefficient is greater than or equal to the first threshold and less than a second threshold, the filtering window is set to a second window length and the polynomial order is set to a second order; when the evaluation coefficient is greater than or equal to the second threshold, the filtering window is set to a third window length and the polynomial order is set to a third order, wherein the first window length is less than the second window length, the second window length is less than the third window length, the first order is greater than the second order, and the second order is greater than the third order.

[0101] Specifically, a first threshold is preset. Second threshold The signal characteristic evaluation coefficient α is based on... =( Normalized value) × ( (Normalized value) calculation. Switching rules are as follows:

[0102] when At that time, the filter window length was set to the first window length of 11 points, and the polynomial order was the first order of 3.

[0103] when At that time, the filter window length was set to the second window length of 21 points, and the polynomial order was set to the second order of 2.

[0104] when At that time, the filter window length was set to the third window length of 31 points, and the polynomial order was set to the third order of 1.

[0105] Wherein, 11 < 21 < 31 (that is, the length of the first window is less than the length of the second window, and the length of the second window is less than the length of the third window), 3 > 2 > 1 (that is, the first order is greater than the second order, and the second order is greater than the third order), to ensure that a small window is used for high fidelity when the signal changes drastically, and a large window is used for strong smoothing when the signal changes smoothly.

[0106] By publicly disclosing the specific values ​​of the three parameter levels and their corresponding thresholds with the evaluation coefficient α, the filtering algorithm can adapt to signal characteristics under different atmospheric conditions without manual debugging, thus improving the robustness and automation of the algorithm.

[0107] Optionally, the upper boundary of the optimal matching region is defined as the upper limit of the effective detection height of the second echo signal, and the lower boundary is the lowest height at which the first echo signal completely enters the telescope's field of view. The step-by-step search to determine the overlapping height interval with the highest correlation and smallest signal strength difference between the first and second echo signals, and marking this overlapping height interval as the optimal matching region, includes: starting from the lower boundary, gradually raising the matching interval by a fixed length according to a preset step size until the upper boundary is reached; calculating the correlation coefficient and the average value of the absolute value of the signal strength difference between the first and second echo signals in each matching interval; selecting the overlapping height interval with the smallest screening coefficient (the correlation coefficient being greater than a first preset threshold and determined by the average value of the absolute value of the signal strength difference and the correlation coefficient), and marking this overlapping height interval as the optimal matching region.

[0108] Specifically, define the upper and lower boundaries: upper boundary This represents the upper limit of the effective detection altitude for the second echo signal (narrow field of view). It is determined through pre-testing: the signal-to-noise ratio (SNR) is calculated point-by-point from low altitude upwards; the altitude at which the SNR first drops below 3 is the upper limit. In this embodiment, the length is 5000 meters. Lower boundary. This is the lowest altitude at which the first echo signal (wide field of view) completely enters the telescope's field of view. Calculated based on the telescope's geometric overlap factor, the altitude is defined as the altitude at which the overlap factor ≥ 0.99. In this embodiment, 200 meters is used.

[0109] Step search: Set a fixed length for the overlap height interval meters. From the starting height Starting with meters, in steps The range gradually increases until the upper boundary of the range is reached. Indexed by serial number Indicates the first The interval obtained from the second rise ( in (This represents the total number of candidate intervals). The starting height (lower boundary) of each candidate interval is denoted as... , .

[0110] The upper boundary of this interval is .when The search is stopped when the time is right. In this embodiment, There are a total of 49 candidate intervals.

[0111] Calculate the correlation coefficient CC for each interval:

[0112] For candidate intervals sampling points ( =500 / 7.5≈67 points), the signal strength in the first field of view detection area is The signal strength in the second field-of-view detection area is , The formula for calculating the correlation coefficient is:

[0113]

[0114] in: Within the interval The arithmetic mean; Within the interval The arithmetic mean of the two signals. The CC value ranges from [-1, 1], and the closer it is to 1, the stronger the positive correlation between the two signals.

[0115] Calculate the average strength difference : .

[0116] Where M is the total number of sampling points within the interval; It represents the absolute difference between the two signal intensities at the i-th sampling point;

[0117] Calculate the screening coefficient : .

[0118] Screening coefficient This comprehensively reflects the consistency of amplitude ( (the smaller the better) and correlation ( The bigger the better. The smaller the value, the better the match.

[0119] Select the best matching region:

[0120] Require The first preset threshold is 0.95, and The smallest candidate interval is taken as the best matching region, and the lower boundary of the best matching region is recorded. and upper boundary .

[0121] By comprehensively evaluating correlation and amplitude consistency, the optimal overlapping area is objectively selected, replacing the traditional subjective method of manual point selection, thus improving the scientific nature and repeatability of the matching.

[0122] Optionally, if there are clouds in the detection path, the cloud base location is taken as the upper boundary of the best matching area.

[0123] During lidar detection, the attenuation coefficient and gradient change of the backscattered signal are calculated in real time. The presence of clouds is determined when both of the following conditions are met:

[0124] Condition 1: The signal strength suddenly increases within a certain altitude range (more than 200 meters continuously), and its peak value exceeds three times the average value of the background signal;

[0125] Condition 2: The signal strength above this interval decreases sharply, with an attenuation rate greater than -10dB / km.

[0126] Once cloud cover is detected, the system automatically detects the cloud base. The cloud base is defined as the initial altitude at which the signal strength begins to rise significantly from the low-altitude plateau, denoted as [missing information]. .

[0127] At this point, the original upper boundary is no longer used. Instead, As the upper boundary of the best matching region. If Below the lower boundary If the match fails, the current matching and fusion process will be cancelled, processing will stop, and an alarm will be output.

[0128] When clouds obscure the signal, the signal above the clouds is severely attenuated, and the signal-to-noise ratio cannot meet the detection requirements. Adjusting the upper boundary to the cloud base position ensures that the matching area is always located in the reliable signal range below the clouds, avoiding false matches caused by cloud interference and improving the algorithm's adaptability to complex atmospheric conditions.

[0129] Optionally, the fusion of the first echo signal and the second echo signal using different fusion strategies includes: for the high-altitude segment, directly using the second echo signal as the inversion data source for that segment; for the fusion segment, using an orthogonal regression fitting method to establish a linear mapping relationship between the first echo signal and the second echo signal, and based on this linear mapping relationship, converting the first echo signal within the fusion segment into data matching the second echo signal; for the near-ground segment, directly using the linear mapping relationship obtained by fitting the fusion segment to map the first echo signal into an equivalent signal for the near-ground segment.

[0130] Let the lower boundary of the best matching region be... (For example, 250 meters), the upper boundary is (For example, 450 meters). The overall detection altitude of 0–10 km is divided into three segments:

[0131] (1) High-altitude section ( ):

[0132] Directly using the second echo signal As an inversion data source.

[0133] (2) Fusion segment ( ):

[0134] The first echo signal was established using orthogonal regression fitting. With the second echo signal The linear mapping relationship between them is expressed by the following linear equation: .

[0135] in: The first echo signal strength (independent variable); The second echo signal strength (dependent variable) is the signal strength. The slope of the fitted line (dimensionless). The intercept of the fitted line (in units of the same as signal strength).

[0136] The optimal parameters are solved iteratively using the Levenberg-Marquardt algorithm with all sampling point pairs within the fusion segment. and After solving, the first echo signal within the fusion segment is converted into a fused signal. : .

[0137] (3) Near the site ( ):

[0138] The same linear equation obtained from the fusion segment is directly used to map the first echo signal into the equivalent signal of the near-ground segment. : .

[0139] The above process describes the clear boundaries and processing methods of the three-stage fusion strategy. The fusion stage establishes a mapping through orthogonal regression, ensuring a smooth transition between the two signals in the transition zone. The near-ground stage uses the same mapping to avoid inconsistencies caused by refitting. The high-altitude stage retains the original narrow field-of-view signal, taking advantage of its high signal-to-noise ratio. This strategy makes the entire profile after fusion physically uniform.

[0140] Optionally, in the fusion segment, the Levenberg-Marquardt algorithm is used to iteratively solve for the optimal linear parameters of the orthogonal regression fit, and the fitting validity is judged based on the root mean square value of the orthogonal residuals. When the root mean square value of the orthogonal residuals is less than a second preset threshold, the optimal linear parameters are output.

[0141] Within the fusion section, there are a total of There are 1 sampling point, and each point corresponds to a pair of signal values. ,in The first echo signal strength, This represents the intensity of the second echo signal. The goal of orthogonal regression is to find a straight line. This minimizes the sum of the squared orthogonal distances from all points to the line.

[0142] The Levenberg-Marquardt algorithm is used for iterative solution:

[0143] Initialization parameters: Set the initial slope Initial intercept Set the iteration termination condition as: root mean square of orthogonal residuals. ,in The second preset threshold is used in this embodiment. Or the number of iterations exceeds 50.

[0144] Calculate the root mean square of orthogonal residuals (ORMS): for the th A point, whose distance to the line orthogonal distance for: .

[0145] in: For the first The orthogonal distance (perpendicular distance) from each sampling point to the fitted line, with the same unit as the signal intensity. For the first The second echo signal strength at each sampling point For the first The first echo signal strength at each sampling point The slope of the fitted line, The intercept of the fitted line; Let be the normalization factor for the orthogonal distance, so that It reflects the vertical distance rather than the vertical projected distance.

[0146] Then the root mean square of the orthogonal residuals for: .

[0147] in: This represents the total number of sampling points within the fusion segment. For the first The square of the orthogonal distance between points To extract the square root, make and They have the same dimensions.

[0148] Iterative parameter updates: The Levenberg-Marquardt algorithm calculates the objective function in each iteration ( )right and The Jacobian matrix, combined with the damping factor, is used to adjust the parameter update step size, gradually reducing the root mean square of the orthogonal residuals. .

[0149] Convergence is determined when... When the iteration stops, output the current value. and As the optimal linear parameter. If after 50 iterations... If the value still does not drop below 0.05, output the parameters of the last iteration and issue a fitting quality alarm.

[0150] The Levenberg-Marquardt algorithm ensures robustness in iterative convergence. Compared to ordinary least squares, orthogonal regression considers both... and The measurement error is more consistent with the actual physical situation that both field-of-view signals have noise, and the fitted mapping relationship is more realistic and reliable.

[0151] Optionally, after the signal conversion of the fusion segment is completed, a distance weighting factor is introduced to smooth the signal values ​​at the junction of the fusion segment and the high-altitude segment and the near-ground segment.

[0152] Specifically, after the signal conversion in the fusion segment is completed, although orthogonal regression fitting is used within the fusion segment and different signal sources are used for the upper-air segment and the near-ground segment, discontinuities may still occur at the boundaries due to fitting residuals or slight fluctuations in the original signal. Therefore, a distance weighting factor is introduced for correction.

[0153] Define the length of the transition region as Meters (approximately 7 sampling points). At the upper boundary of the fusion segment. Nearby, from arrive Within the range, and at the lower boundary of the fusion segment. Nearby, from arrive Within the range, perform weighted smoothing.

[0154] Weighting factors Definition:

[0155] Near the upper boundary: for distance ,definition: .

[0156] Right now The value decreases linearly from 1 at the lower end of the transition region to 0 at the upper end.

[0157] Near the lower boundary: for distance ,definition: .

[0158] Right now The transition zone increases linearly from 0 at the lower end to 1 at the upper end.

[0159] in: The current distance coordinates, The upper boundary of the best matching region. The lower boundary of the best matching region. The length of the transition zone is 50 meters in this embodiment.

[0160] Calculation of the final fused signal:

[0161] Near the upper boundary: .

[0162] in Fitted values ​​for the fusion segment This is the original second echo signal in the high-altitude segment.

[0163] Near the lower boundary: .

[0164] Where P1_mapped(R) is the direct mapping value of the near-ground segment (i.e. ).

[0165] The above process discloses a simple and effective boundary smoothing correction method, defines a transition zone length of 50 meters, and provides a weighting factor. The specific formula for linear variation with distance and the calculation formula for the final fused signal are presented. Through weighted averaging, the signal values ​​on both sides of the boundary transition smoothly, completely eliminating minor abrupt changes caused by fitting residuals or signal fluctuations, ensuring the first-order continuity of the entire fused signal profile. This correction does not affect signal characteristics far from the boundary, but only acts on the local transition region, ensuring both smoothness and the authenticity of the original data.

[0166] Reference Figure 2 This specification also provides a structural diagram of a lidar signal matching, fusion, and inversion device, comprising:

[0167] The signal acquisition module 201 is used to acquire the pre-processed first echo signal and the second echo signal respectively. The first echo signal is the signal obtained by using a wide field-of-view receiving telescope to detect the first field-of-view detection area, and the second echo signal is the signal obtained by using a narrow field-of-view receiving telescope to detect the second field-of-view detection area.

[0168] The optimal matching region determination module 202 is used to determine the overlapping height interval between the first echo signal and the second echo signal with the highest correlation and the smallest signal strength difference within a preset height range through step search, and to mark the overlapping height interval as the optimal matching region.

[0169] The signal fusion module 203 is used to divide the global detection altitude into a high-altitude segment, a fusion segment, and a near-ground segment based on the upper and lower boundaries of the optimal matching area, and to fuse the first echo signal and the second echo signal using different fusion strategies to obtain a fused signal.

[0170] The inversion module 204 is used to invert the concentration profile of the characteristic pollutant within the global detection height range based on the fused signal.

[0171] Regarding the apparatus in the above embodiments, the process of performing each step has been described in detail in the embodiments of the method, and will not be elaborated here.

[0172] Based on the same inventive concept, embodiments of this specification also provide an electronic device.

[0173] The following describes embodiments of the electronic device of the present invention, which can be considered as specific implementations of the methods and apparatus embodiments of the present invention described above. Details described in the embodiments of the electronic device of the present invention should be considered as supplements to the methods or apparatus embodiments described above; details not disclosed in the embodiments of the electronic device of the present invention can be implemented with reference to the methods or apparatus embodiments described above.

[0174] Reference Figure 3This is a schematic diagram of an electronic device provided as an embodiment of this specification. Refer to the following... Figure 3 The electronic device 300 according to this embodiment of the present invention will be described. Figure 3 The electronic device 300 shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0175] like Figure 3 As shown, the electronic device 300 is presented in the form of a general-purpose computing device. The components of the electronic device 300 may include, but are not limited to: at least one processing unit 310, at least one storage unit 320, a bus 330 connecting different device components (including storage unit 320 and processing unit 310), a display unit 340, etc.

[0176] The storage unit stores program code that can be executed by the processing unit 310, causing the processing unit 310 to perform the steps described in the processing method section of this specification according to various exemplary embodiments of the present invention. For example, the processing unit 310 can perform, for example... Figure 1 The steps are shown.

[0177] The storage unit 320 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 3201 and / or a cache storage unit 3202, and may further include a read-only memory unit (ROM) 3203.

[0178] The storage unit 320 may also include a program / utility 3204 having a set (at least one) of program modules 3205, such program modules 3205 including but not limited to: operating devices, one or more application programs, other program modules and program data, each or some combination of these examples may include an implementation of a network environment.

[0179] Bus 330 can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the various bus structures.

[0180] Electronic device 300 can also communicate with one or more external devices 400 (e.g., keyboard, pointing device, Bluetooth device, etc.), and with one or more devices that enable a user to interact with electronic device 300, and / or with any device that enables electronic device 300 to communicate with one or more other computing devices (e.g., router, modem, etc.). This communication can be performed via input / output (I / O) interface 350. Furthermore, electronic device 300 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 360. Network adapter 360 can communicate with other modules of electronic device 300 via bus 330. It should be understood that, although... Figure 3 As not shown, other hardware and / or software modules may be used in conjunction with electronic device 300, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID devices, tape drives, and data backup storage devices.

[0181] Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described in this invention can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this invention can be embodied in the form of a software product, which can be stored in a computer-readable storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, or network device, etc.) to execute the method described above according to this invention. When the computer instructions are executed by a data processing device, the computer-readable medium is able to implement the method described above, i.e., as follows: Figure 1 The method shown.

[0182] Reference Figure 4 This is a schematic diagram of a computer-readable medium provided for embodiments of this specification.

[0183] accomplish Figure 1The computer instructions of the method shown can be stored on one or more computer-readable media. A computer-readable medium can be a readable signal medium or a readable storage medium. A readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor device, or any combination thereof. More specific examples (a non-exhaustive list) of readable storage media include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0184] The computer-readable storage medium may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The readable storage medium may also be any readable medium other than a readable storage medium, capable of transmitting, propagating, or transmitting a program for use by or in connection with an instruction execution device, apparatus, or apparatus. The program code contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0185] Program code for performing the operations of this invention can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0186] In summary, the present invention can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that in practice, general-purpose data processing devices such as microprocessors or digital signal processors (DSPs) can be used to implement some or all of the functions of some or all of the components according to the embodiments of the present invention. The present invention can also be implemented as a device or apparatus program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such programs implementing the present invention can be stored on a computer-readable medium or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

[0187] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the present invention is not inherently related to any specific computer, virtual device, or electronic device, and various general-purpose devices can also implement the present invention. The above descriptions are merely specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0188] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0189] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for matching, fusing, and inverting lidar signals, characterized in that, include: The first echo signal and the second echo signal after preprocessing are acquired respectively. The first echo signal is the signal obtained by using a wide field-of-view receiving telescope to detect the first field-of-view detection area, and the second echo signal is the signal obtained by using a narrow field-of-view receiving telescope to detect the second field-of-view detection area. Within a preset height range, the overlapping height interval between the first echo signal and the second echo signal with the highest correlation and the smallest signal strength difference is determined by step search, and the overlapping height interval is marked as the best matching area. Based on the upper and lower boundaries of the optimal matching area, the overall detection altitude is divided into a high-altitude segment, a fusion segment, and a near-ground segment. Different fusion strategies are used to fuse the first echo signal and the second echo signal to obtain a fused signal. The concentration profile of the characteristic pollutant within the entire detection height range is obtained based on the fused signal inversion.

2. The method according to claim 1, characterized in that, The first field of view detection area is the area used to detect near-ground range signals, and the second field of view detection area is the area used to detect high-altitude range signals; the preprocessing includes: performing distance correction, background noise suppression, and validity discrimination on the first echo signal and the second echo signal, and performing adaptive cascaded Savitzky-Golay filtering noise reduction processing on the signal that passes the validity discrimination.

3. The method according to claim 2, characterized in that, The signal validity determination includes: Determine whether there is electromagnetic interference and inter-channel crosstalk in the first echo signal and the second echo signal. If so, mark the corresponding echo signal as invalid data.

4. The method according to claim 2, characterized in that, The signal validity determination also includes: Calculate the signal-to-noise ratio (SNR) of the first echo signal and the second echo signal at each distance, and determine the signal with an SNR greater than a preset threshold as valid data.

5. The method according to claim 2, characterized in that, The signal validity determination also includes: real-time monitoring of the received signal strength of the first echo signal and the second echo signal; The gain parameter of the photodetector used to convert optical signals into electrical signals is dynamically adjusted or an optical attenuator is inserted based on the received signal strength, so that the received signal strength is always below a preset saturation threshold.

6. The method according to claim 2, characterized in that, The adaptive cascaded Savitzky-Golay filtering noise reduction process includes: The window signal-to-noise ratio and gradient rate of change of the echo signal determined to be valid are calculated in real time; the gradient rate of change is obtained by calculating the mean of the absolute values ​​of the differential absorption of adjacent sampling points of the echo signal determined to be valid. The signal characteristics of the effective echo signal are dynamically evaluated based on the window signal-to-noise ratio and gradient change rate, and the window length and polynomial order of the Savitzky-Golay filter are adaptively set according to the evaluation results. Return to the process of performing real-time calculations of the window signal-to-noise ratio and gradient rate of change of the valid echo signal until the signal-to-noise ratio increment of the valid echo signal at the maximum detection distance is less than a preset threshold.

7. The method according to claim 1, characterized in that, The upper boundary of the optimal matching region is defined as the upper limit of the effective detection height of the second echo signal, and the lower boundary is the lowest height at which the first echo signal completely enters the telescope's field of view; The step-by-step search determines the overlap height interval between the first and second echo signals, where the correlation is highest and the signal strength difference is smallest. This overlap height interval is then marked as the optimal matching region. Starting from the lower boundary, the matching interval is gradually raised by a fixed length according to a preset step size until the upper boundary is reached. The correlation coefficient between the first echo signal and the second echo signal, as well as the average value of the absolute value of the signal strength difference, are calculated in each matching interval. Select the overlap height interval with the smallest screening coefficient that is greater than the first preset threshold and is determined by the average value of the absolute value of the signal strength difference and the correlation coefficient, and mark the overlap height interval as the best matching region.

8. The method according to claim 1, characterized in that, The method of fusing the first echo signal and the second echo signal using different fusion strategies includes: For the high-altitude segment, the second echo signal is directly used as the inversion data source for that segment; For the fusion segment, an orthogonal regression fitting method is used to establish a linear mapping relationship between the first echo signal and the second echo signal, and based on this linear mapping relationship, the first echo signal in the fusion segment is converted into data that matches the second echo signal; For the near-ground segment, the linear mapping relationship obtained by fitting the fusion segment is directly used to map the first echo signal into the equivalent signal of the near-ground segment.

9. The method according to claim 8, characterized in that, After the signal conversion of the fusion segment is completed, a distance weighting factor is introduced to smooth the signal values ​​at the junction of the fusion segment and the high-altitude segment and the near-ground segment.

10. A laser radar signal matching, fusion, and inversion device, characterized in that, include: The signal acquisition module is used to acquire the pre-processed first echo signal and the second echo signal respectively. The first echo signal is the signal obtained by using a wide field-of-view receiving telescope to detect the first field-of-view detection area, and the second echo signal is the signal obtained by using a narrow field-of-view receiving telescope to detect the second field-of-view detection area. The optimal matching region determination module is used to determine the overlapping height interval between the first echo signal and the second echo signal with the highest correlation and the smallest signal strength difference within a preset height range through step search, and to mark the overlapping height interval as the optimal matching region. The signal fusion module is used to divide the global detection altitude into a high-altitude segment, a fusion segment, and a near-ground segment based on the upper and lower boundaries of the optimal matching area, and to fuse the first echo signal and the second echo signal using different fusion strategies to obtain a fused signal. The inversion module is used to invert the concentration profile of the characteristic pollutant within the global detection height range based on the fused signal.

Images