Multi-target recognition method in complex traffic environment and electronic device

Abstract

The disclosure provides a multi-target recognition method and an electronic device in a complex traffic environment. The accuracy of target recognition is improved. It includes: for any one frame of road point cloud data in the acquired road point cloud data of a specified number of frames, projecting the road point cloud data in the radial distance of the millimeter wave radar to obtain the high resolution range profile (HRRP) of the road point cloud data, performing point cloud feature extraction on the road point cloud data to obtain the point cloud feature of the road point cloud data; according to the point cloud feature of the road point cloud data and the HRRP geometric feature and HRRP power spectrum feature obtained by performing feature extraction on the HRRP of the road point cloud data, obtain the point cloud feature set of the road point cloud data; according to the point cloud feature set of the road point cloud data of the specified number of frames, obtain the point cloud feature sequence; input the point cloud feature sequence into the pre-trained target classification model to obtain the category of each target object in the road point cloud data of the specified number of frames.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent transportation, in particular to a multi-target recognition method in a complex traffic environment and an electronic device. BACKGROUND

[0002] The complex traffic environment refers to a mixed traffic composed of motor vehicles, non-motor vehicles and pedestrians. It seriously affects the road traffic order, reduces the road traffic capacity and increases the difficulty of road traffic management.

[0003] Based on traffic road vehicle recognition, traffic control, real-time monitoring and statistics of the position, speed and quantity of various types of vehicles on the road are realized, and a solution to prevent traffic accidents from the root is an effective way to quickly improve the road traffic environment and improve traffic safety.

[0004] In the prior art, the traffic road vehicle recognition is realized based on the classification result obtained from the point cloud data generated by single detection of the target. However, in the actual road target classification task, the classification effect of this method is easily affected by abnormal detection results, and the accuracy of target recognition in a crowded and complex traffic environment is low only by relying on single frame of point cloud data. SUMMARY

[0005] In the exemplary embodiments of the present disclosure, a multi-target recognition method in a complex traffic environment and an electronic device are provided to improve the accuracy of target recognition in a complex traffic environment.

[0006] The first aspect of the present disclosure provides a multi-target recognition method in a complex traffic environment, the method comprising:

[0007] Every specified time interval, a specified number of road point cloud data is obtained, wherein the road point cloud data is obtained by detecting the target road using a millimeter wave radar;

[0008] For any one frame of road point cloud data, the road point cloud data is projected in the radial distance of the millimeter wave radar to obtain a high-resolution range profile (HRRP) of the road point cloud data, and point cloud feature extraction is performed on the road point cloud data to obtain respective point cloud features of the road point cloud data; and

[0009] According to the point cloud features, HRRP geometric features and HRRP power spectrum features of the road point cloud data, a point cloud feature set of the road point cloud data is obtained, wherein the HRRP geometric features and the HRRP power spectrum features are obtained by feature extraction on the HRRP of the road point cloud data;

[0010] According to the point cloud feature sets of the specified number of road point cloud data, a point cloud feature sequence is obtained.

[0011] inputting the point cloud feature sequence into a pre-trained target classification model to obtain the category of each target object in the road point cloud data of the specified frame number.

[0012] In the embodiment, the point cloud feature set of the single-frame road point cloud data is determined by extracting the point cloud feature, the HRRP geometric feature and the HRRP power spectrum feature of the single-frame road point cloud data, and then the point cloud feature sequence is obtained by combining the point cloud feature sets of the multiple frames of road point cloud data. The category of each target object is obtained by target recognition based on the point cloud feature sequence. Thus, in the application embodiment, multiple features in the single-frame road point cloud data are extracted to accurately obtain the point cloud feature set of each target object, and the point cloud feature sequence of multiple frames of high dimension is obtained by fusing the point cloud feature sets of multiple frames of road point cloud data in the time sequence, so that more detailed information of each target object in the complex traffic environment can be captured to improve the accuracy of multi-target recognition in the complex traffic environment.

[0013] In one embodiment, the road point cloud data includes the position coordinates, radial velocity and echo intensity of each target point; and any one point cloud feature includes at least one of the point cloud number, the horizontal position coordinate of the point cloud centroid, the vertical position coordinate of the point cloud centroid, the ratio of the horizontal and vertical coordinates of the point cloud centroid, the relative radar distance of the point cloud centroid, the radial velocity of the point cloud centroid, the radial velocity variance of the point cloud centroid, the radial velocity change difference of the point cloud centroid, the minimum value of the point cloud echo intensity, the maximum value of the point cloud echo intensity, the mean value of the point cloud echo intensity, the variance of the point cloud echo intensity, the difference of the horizontal position coordinates of the point cloud, the difference of the vertical position coordinates of the point cloud, the variance of the horizontal position coordinates of the point cloud, the variance of the vertical position coordinates of the point cloud and the point cloud density.

[0014] The point cloud feature extraction on the road point cloud data to obtain the point cloud feature of the road point cloud data includes:

[0015] The total number of each target point in the road point cloud data is determined as the point cloud number; and / or,

[0016] The mean value of the horizontal position coordinates of each target point in the road point cloud data is determined as the horizontal position coordinate of the point cloud centroid; and / or,

[0017] The mean value of the vertical position coordinates of each target point in the road point cloud data is determined as the vertical position coordinate of the point cloud centroid; and / or,

[0018] The horizontal position coordinate of the point cloud centroid is divided by the vertical position coordinate of the point cloud centroid to obtain the ratio of the horizontal and vertical coordinates of the point cloud centroid; and / or,

[0019] obtaining the point cloud centroid relative radar distance according to the horizontal position coordinate of the point cloud centroid and the vertical position coordinate of the point cloud centroid; and / or,

[0020] determining the point cloud centroid radial velocity as the average value of the radial velocities of the target points in the road point cloud data; and / or,

[0021] obtaining the point cloud centroid radial velocity variance according to the radial velocities of the target points in the road point cloud data and the point cloud centroid radial velocity; and / or,

[0022] determining the point cloud centroid radial velocity change difference as the difference between the radial velocity of the target point with the maximum value and the radial velocity of the target point with the minimum value in the road point cloud data; and / or,

[0023] determining the point cloud echo intensity minimum value as the echo intensity of the target point with the minimum value in the road point cloud data; and / or,

[0024] determining the point cloud echo intensity maximum value as the echo intensity of the target point with the maximum value in the road point cloud data; and / or,

[0025] determining the point cloud echo intensity mean value as the mean value of the echo intensities of the target points in the road point cloud data;

[0026] obtaining the point cloud echo intensity variance according to the echo intensities of the target points in the road point cloud data and the point cloud echo intensity mean value;

[0027] obtaining the point cloud horizontal position coordinate difference by subtracting the horizontal position coordinate of the target point with the minimum value from the horizontal position coordinate of the target point with the maximum value in the road point cloud data;

[0028] obtaining the point cloud vertical position coordinate difference by subtracting the vertical position coordinate of the target point with the minimum value from the vertical position coordinate of the target point with the maximum value in the road point cloud data;

[0029] obtaining the point cloud horizontal position coordinate variance according to the horizontal position coordinates of the target points in the road point cloud data and the horizontal position coordinate of the point cloud centroid;

[0030] obtaining the point cloud vertical position coordinate variance according to the vertical position coordinates of the target points in the road point cloud data and the vertical position coordinate of the point cloud centroid;

[0031] obtaining the point cloud density according to the point cloud horizontal position coordinate difference, the point cloud vertical position coordinate difference and the point cloud quantity.

[0032] In one embodiment, the determining the mean value of the lateral position coordinates of each target point in the road point cloud data as the lateral position coordinate of the point cloud centroid comprises:

[0033] The lateral position coordinate of the point cloud centroid is obtained by the following formula:

[0034]

[0035] wherein, is the lateral position coordinate of the point cloud centroid, x i is the lateral position coordinate of target point i in the road point cloud data, i∈[1, N], N is the total number of target points in the road point cloud data;

[0036] The determining the mean value of the longitudinal position coordinates of each target point in the road point cloud data as the longitudinal position coordinate of the point cloud centroid comprises:

[0037] The longitudinal position coordinate of the point cloud centroid is obtained by the following formula:

[0038]

[0039] wherein, is the longitudinal position coordinate of the point cloud centroid, y i is the longitudinal position coordinate of target point i in the road point cloud data;

[0040] The dividing the lateral position coordinate of the point cloud centroid by the longitudinal position coordinate of the point cloud centroid to obtain the point cloud centroid lateral-longitudinal coordinate ratio comprises:

[0041] The point cloud centroid lateral-longitudinal coordinate ratio is obtained by the following formula:

[0042]

[0043] wherein, σ is the point cloud centroid lateral-longitudinal coordinate ratio;

[0044] The obtaining the point cloud centroid relative radar distance according to the lateral position coordinate of the point cloud centroid and the longitudinal position coordinate of the point cloud centroid comprises:

[0045] The point cloud centroid relative radar distance is obtained by the following formula:

[0046]

[0047] wherein, R is the point cloud centroid relative radar distance;

[0048] The determining the mean value of the radial velocities of each target point in the road point cloud data as the point cloud centroid radial velocity comprises:

[0049] The point cloud centroid radial velocity is obtained by the following manner:

[0050]

[0051] wherein, v i is the radial velocity of the target point i in the road point cloud data;

[0052] The point cloud centroid radial velocity variance is obtained according to the radial velocity of each target point in the road point cloud data and the point cloud centroid radial velocity, comprising:

[0053] The point cloud centroid radial velocity variance is obtained by the following formula:

[0054]

[0055] wherein, δ v v i is the radial velocity of the target point i in the road point cloud data, is the point cloud centroid radial velocity;

[0056] The difference between the radial velocity of the target point with the largest value in the road point cloud data and the radial velocity of the target point with the smallest value is determined as the point cloud centroid radial velocity variation difference, comprising:

[0057] The point cloud centroid radial velocity variation difference is obtained by the following formula:

[0058] a v = v max - v min ;

[0059] wherein, a v v max is the radial velocity of the target point with the largest value in the road point cloud data, v min is the radial velocity of the target point with the smallest value in the road point cloud data;

[0060] The mean value of the echo intensity of each target point in the road point cloud data is determined as the point cloud echo intensity mean value, comprising:

[0061] The point cloud echo intensity mean value is obtained by the following formula:

[0062]

[0063] wherein, is the point cloud echo intensity mean value, dB ian echo intensity of a target point i in the road point cloud data;

[0064] The point cloud echo intensity variance is obtained according to the echo intensity of each target point in the road point cloud data and the point cloud echo intensity mean value, and the method comprises the following steps:

[0065] The point cloud echo intensity variance is obtained by the following formula:

[0066]

[0067] Wherein, δ dB is the point cloud echo intensity variance, dB i is an echo intensity of a target point i in the road point cloud data, dB is the point cloud echo intensity mean value;

[0068] The point cloud horizontal position coordinate difference is obtained by subtracting the horizontal position coordinate of the target point with the minimum value from the horizontal position coordinate of the target point with the maximum value in the road point cloud data, and the method comprises the following steps:

[0069] a x = x max - x min ;

[0070] Wherein, a x is the point cloud horizontal position coordinate difference, x max is the horizontal position coordinate of the target point with the maximum value in the road point cloud data, and x min is the horizontal position coordinate of the target point with the minimum value in the road point cloud data;

[0071] The point cloud vertical position coordinate difference is obtained by subtracting the vertical position coordinate of the target point with the minimum value from the vertical position coordinate of the target point with the maximum value in the road point cloud data, and the method comprises the following steps:

[0072] The point cloud vertical position coordinate difference is obtained by the following formula:

[0073] a y = y max - y min ;

[0074] Wherein, a y is the point cloud vertical position coordinate difference, y max is the vertical position coordinate of the target point with the maximum value in the road point cloud data, and y min is the vertical position coordinate of the target point with the minimum value in the road point cloud data;

[0075] The point cloud horizontal position coordinate variance is obtained according to the horizontal position coordinate of each target point in the road point cloud data and the horizontal position coordinate of the point cloud centroid, and the method comprises the following steps:

[0076] The point cloud horizontal position coordinate variance is obtained by the following formula:

[0077]

[0078] wherein, δ x is the point cloud horizontal position coordinate variance, x i is the horizontal position coordinate of target point i in the road point cloud data, is the horizontal position coordinate of the point cloud centroid;

[0079] The point cloud vertical position coordinate variance is obtained according to the vertical position coordinates of each target point in the road point cloud data and the vertical position coordinate of the point cloud centroid, and the point cloud vertical position coordinate variance comprises:

[0080] The point cloud vertical position coordinate variance is obtained by the following formula:

[0081]

[0082] wherein, δ y is the point cloud vertical position coordinate variance, y i is the vertical position coordinate of target point i in the road point cloud data, is the vertical position coordinate of the point cloud centroid;

[0083] The point cloud density is obtained according to the point cloud horizontal position coordinate difference, the point cloud vertical position coordinate difference and the point cloud quantity, and the point cloud density comprises:

[0084] The point cloud density is obtained by the following formula:

[0085]

[0086] wherein, ρ is the point cloud density, α x is the point cloud horizontal position coordinate difference, α y is the point cloud vertical position coordinate difference.

[0087] In one embodiment, the HRRP of the road point cloud data comprises the amplitude of each target point in the road point cloud data;

[0088] The HRRP geometric features comprise at least one of the equivalent scattering center quantity, the equivalent target size, the HRRP signal entropy, the HRRP signal standard deviation, the HRRP signal skewness, the HRRP signal kurtosis, the HRRP signal echo energy, the HRRP signal energy ratio, the HRRP signal skewness and the HRRP signal kurtosis;

[0089] The equivalent scattering center quantity is obtained by the following formula:

[0090]

[0091] wherein, Num sca is the number of equivalent scattering centers, y(i) is the amplitude of target point i, m is the average of the amplitudes of each target point in the road point cloud data, N is the total number of each target point in the road point cloud data, when y(i)-m≥0, ε(y(i)-m)=1, when y(i)-m<0, ε(y(i)-m)=0;

[0092] The equivalent target size is obtained by the following formula:

[0093] The equivalent target size is obtained according to the position coordinates of the target point with the maximum amplitude and the position coordinates of the target point with the minimum amplitude in the HRRP;

[0094] The HRRP signal entropy is obtained by the following formula:

[0095]

[0096] wherein, Entropy is the HRRP signal entropy;

[0097] The HRRP signal standard deviation is obtained by the following formula:

[0098]

[0099] wherein, Std is the HRRP signal standard deviation;

[0100] The HRRP signal deviation is obtained by the following formula:

[0101]

[0102] Deviation is the HRRP signal deviation;

[0103] The HRRP signal irregularity is obtained by the following formula:

[0104]

[0105] wherein, Irr is the HRRP signal irregularity, y(i-1) is the amplitude of the target point i-1 before the target point i, y(i+1) is the amplitude of the target point i+1 after the target point i;

[0106] The HRRP signal echo energy is obtained by the following formula:

[0107]

[0108] wherein, TP is the HRRP signal echo energy;

[0109] The HRRP signal energy ratio is obtained by the following manner:

[0110] The HRRP signal energy ratio is obtained by the following manner:

[0111] The HRRP signal skewness is obtained by the following formula:

[0112]

[0113] Wherein, Skewness is the HRRP signal skewness, and μ is the average of the absolute values of the amplitudes of the target points.

[0114] The HRRP signal kurtosis is obtained by the following formula:

[0115]

[0116] Wherein, kurtosis is the HRRP signal kurtosis.

[0117] In an embodiment, the HRRP power spectrum feature of the road point cloud data is obtained by the following manner:

[0118] For the amplitude of any target point in the HRRP of the road point cloud data, the amplitude of the target point is subjected to Fourier transform to obtain the frequency domain amplitude of the target point; and

[0119] The square of the frequency domain amplitude of the target point is determined as the power spectrum of the target point.

[0120] The power spectrum of each target point in the HRRP of the road point cloud data is determined as the HRRP power spectrum feature of the road point cloud data.

[0121] The second aspect of the present disclosure provides an electronic device, comprising a processor and a memory, the processor and the memory are connected through a bus;

[0122] The memory stores a computer program, and the processor is configured to perform the following operations based on the computer program:

[0123] Every specified time length, a specified number of frames of road point cloud data are obtained, wherein the road point cloud data is obtained by detecting the target road by using a millimeter wave radar;

[0124] The road point cloud data is projected on a radial distance of the millimeter wave radar to obtain a high resolution range profile (HRRP) of the road point cloud data, and a point cloud feature of the road point cloud data is extracted to obtain a point cloud feature of the road point cloud data; and

[0125] The point cloud feature of the road point cloud data, an HRRP geometric feature, and an HRRP power spectrum feature are obtained to obtain a point cloud feature set of the road point cloud data, wherein the HRRP geometric feature and the HRRP power spectrum feature are obtained by extracting features of the HRRP of the road point cloud data;

[0126] The point cloud feature set of the road point cloud data of the specified number of frames is obtained to obtain a point cloud feature sequence.

[0127] The point cloud feature sequence is input into a target classification model that is trained in advance to obtain a category of each target object in the road point cloud data of the specified number of frames.

[0128] In one embodiment, the road point cloud data includes position coordinates, radial velocities, and echo intensities of each target point, and any one point cloud feature includes at least one of a point cloud number, a horizontal position coordinate of a point cloud centroid, a vertical position coordinate of the point cloud centroid, a ratio of the horizontal position coordinate to the vertical position coordinate of the point cloud centroid, a relative radar distance of the point cloud centroid, a radial velocity of the point cloud centroid, a radial velocity variance of the point cloud centroid, a radial velocity change difference of the point cloud centroid, a minimum value of a point cloud echo intensity, a maximum value of the point cloud echo intensity, a mean value of the point cloud echo intensity, a variance of the point cloud echo intensity, a horizontal position coordinate difference of the point cloud, a vertical position coordinate difference of the point cloud, a horizontal position coordinate variance of the point cloud, a vertical position coordinate variance of the point cloud, and a point cloud density.

[0129] The processor performs the point cloud feature extraction on the road point cloud data to obtain the point cloud feature of the road point cloud data, and is specifically configured to:

[0130] The total number of each target point in the road point cloud data is determined as the point cloud number; and / or,

[0131] The mean value of the horizontal position coordinates of each target point in the road point cloud data is determined as the horizontal position coordinate of the point cloud centroid; and / or,

[0132] The mean value of the vertical position coordinates of each target point in the road point cloud data is determined as the vertical position coordinate of the point cloud centroid; and / or,

[0133] The horizontal position coordinate of the point cloud centroid is divided by the vertical position coordinate of the point cloud centroid to obtain the ratio of the horizontal position coordinate to the vertical position coordinate of the point cloud centroid; and / or,

[0134] obtaining the point cloud centroid relative radar distance according to the horizontal position coordinate of the point cloud centroid and the vertical position coordinate of the point cloud centroid; and / or,

[0135] determining the point cloud centroid radial velocity as the average value of the radial velocities of the target points in the road point cloud data; and / or,

[0136] obtaining the point cloud centroid radial velocity variance according to the radial velocities of the target points in the road point cloud data and the point cloud centroid radial velocity; and / or,

[0137] determining the point cloud centroid radial velocity change difference as the difference between the radial velocity of the target point with the maximum value and the radial velocity of the target point with the minimum value in the road point cloud data; and / or,

[0138] determining the point cloud echo intensity minimum value as the echo intensity of the target point with the minimum value in the road point cloud data; and / or,

[0139] determining the point cloud echo intensity maximum value as the echo intensity of the target point with the maximum value in the road point cloud data; and / or,

[0140] determining the point cloud echo intensity mean value as the mean value of the echo intensities of the target points in the road point cloud data;

[0141] obtaining the point cloud echo intensity variance according to the echo intensities of the target points in the road point cloud data and the point cloud echo intensity mean value;

[0142] obtaining the point cloud horizontal position coordinate difference by subtracting the horizontal position coordinate of the target point with the minimum value from the horizontal position coordinate of the target point with the maximum value in the road point cloud data;

[0143] obtaining the point cloud vertical position coordinate difference by subtracting the vertical position coordinate of the target point with the minimum value from the vertical position coordinate of the target point with the maximum value in the road point cloud data;

[0144] obtaining the point cloud horizontal position coordinate variance according to the horizontal position coordinates of the target points in the road point cloud data and the horizontal position coordinate of the point cloud centroid;

[0145] obtaining the point cloud vertical position coordinate variance according to the vertical position coordinates of the target points in the road point cloud data and the vertical position coordinate of the point cloud centroid;

[0146] obtaining the point cloud density according to the point cloud horizontal position coordinate difference, the point cloud vertical position coordinate difference and the point cloud quantity.

[0147] In one embodiment, the processor performs determination of a mean value of the horizontal position coordinates of each target point in the road point cloud data as the horizontal position coordinate of the point cloud centroid, and is specifically configured to:

[0148] The horizontal position coordinate of the point cloud centroid is obtained by the following formula:

[0149]

[0150] wherein, is the horizontal position coordinate of the point cloud centroid, x i is the horizontal position coordinate of the target point i in the road point cloud data, i∈[1, N], N is the total number of target points in the road point cloud data;

[0151] The processor performs determination of a mean value of the vertical position coordinates of each target point in the road point cloud data as the vertical position coordinate of the point cloud centroid, and is specifically configured to:

[0152] The vertical position coordinate of the point cloud centroid is obtained by the following formula:

[0153]

[0154] wherein, is the vertical position coordinate of the point cloud centroid, y i is the vertical position coordinate of the target point i in the road point cloud data;

[0155] The processor performs division of the horizontal position coordinate of the point cloud centroid by the vertical position coordinate of the point cloud centroid to obtain a horizontal-to-vertical coordinate ratio of the point cloud centroid, and is specifically configured to:

[0156] The horizontal-to-vertical coordinate ratio of the point cloud centroid is obtained by the following formula:

[0157]

[0158] wherein, σ is the horizontal-to-vertical coordinate ratio of the point cloud centroid;

[0159] The processor performs determination of the relative radar distance of the point cloud centroid according to the horizontal position coordinate of the point cloud centroid and the vertical position coordinate of the point cloud centroid, and is specifically configured to:

[0160] The relative radar distance of the point cloud centroid is obtained by the following formula:

[0161]

[0162] wherein, R is the relative radar distance of the point cloud centroid;

[0163] The processor is specifically configured to determine the average radial velocity of each target point in the road point cloud data as the centroid radial velocity of the point cloud.

[0164] The radial velocity of the centroid of the point cloud is obtained in the following manner:

[0165]

[0166] in, v is the radial velocity of the centroid of the point cloud. i Let be the radial velocity of target point i in the road point cloud data;

[0167] The processor executes the step of obtaining the variance of the radial velocity of the centroid of the point cloud based on the radial velocity of each target point in the road point cloud data and the radial velocity of the centroid of the point cloud, specifically configured as follows:

[0168] The variance of the radial velocity of the centroid of the point cloud is obtained using the following formula:

[0169]

[0170] Where, δ v Let v be the variance of the radial velocity of the centroid of the point cloud. i The radial velocity of target point i in the road point cloud data. The radial velocity of the centroid of the point cloud;

[0171] The processor executes the step of determining the difference between the radial velocity of the target point with the largest value and the radial velocity of the target point with the smallest value in the road point cloud data as the difference in radial velocity change of the centroid of the point cloud, specifically configured as follows:

[0172] The difference in radial velocity variation of the centroid of the point cloud is obtained using the following formula:

[0173] a v =v max -v min ;

[0174] Among them, a v v represents the difference in radial velocity variation of the centroid of the point cloud. max v is the radial velocity of the target point with the largest value in the road point cloud data. min The radial velocity of the target point with the smallest value in the road point cloud data;

[0175] The processor is specifically configured to determine the mean echo intensity of each target point in the road point cloud data as the mean echo intensity of the point cloud.

[0176] The mean value of the point cloud echo intensity is obtained using the following formula:

[0177]

[0178] wherein, is the mean of the point cloud echo intensity, dB i is the echo intensity of the target point i;

[0179] The processor performs the obtaining of the point cloud echo intensity variance according to the echo intensity of each target point in the road point cloud data and the mean of the point cloud echo intensity, and is specifically configured to:

[0180] The point cloud echo intensity variance is obtained through the following formula:

[0181]

[0182] wherein, δ dB is the point cloud echo intensity variance, dB i is the echo intensity of the target point i in the road point cloud data, is the mean of the point cloud echo intensity;

[0183] The processor performs the obtaining of the point cloud horizontal position coordinate difference by subtracting the horizontal position coordinate of the target point with the maximum value in the road point cloud data from the horizontal position coordinate of the target point with the minimum value, and is specifically configured to:

[0184] a x = x max - x min ;

[0185] wherein, a x is the point cloud horizontal position coordinate difference, x max is the horizontal position coordinate of the target point with the maximum value in the road point cloud data, and x min is the horizontal position coordinate of the target point with the minimum value in the road point cloud data;

[0186] The processor performs the obtaining of the point cloud vertical position coordinate difference by subtracting the vertical position coordinate of the target point with the minimum value in the road point cloud data from the vertical position coordinate of the target point with the maximum value, and is specifically configured to:

[0187] The point cloud vertical position coordinate difference is obtained through the following formula:

[0188] a y = y max - y min ;

[0189] wherein, a y is the point cloud vertical position coordinate difference, y maxis a longitudinal position coordinate of a target point with the largest value in the road point cloud data, y min is a longitudinal position coordinate of a target point with the largest value in the road point cloud data;

[0190] The processor performs the obtaining of the point cloud horizontal position coordinate variance according to the horizontal position coordinates of the target points in the road point cloud data and the horizontal position coordinate of the point cloud centroid, and is specifically configured to:

[0191] The point cloud horizontal position coordinate variance is obtained through the following formula:

[0192]

[0193] wherein, δ x is the point cloud horizontal position coordinate variance, x i is a horizontal position coordinate of a target point i in the road point cloud data, is the horizontal position coordinate of the point cloud centroid;

[0194] The processor performs the obtaining of the point cloud longitudinal position coordinate variance according to the longitudinal position coordinates of the target points in the road point cloud data and the longitudinal position coordinate of the point cloud centroid, and is specifically configured to:

[0195] The point cloud longitudinal position coordinate variance is obtained through the following formula:

[0196]

[0197] wherein, δ y is the point cloud longitudinal position coordinate variance, y i is a longitudinal position coordinate of a target point i in the road point cloud data, is the longitudinal position coordinate of the point cloud centroid;

[0198] The processor performs the obtaining of the point cloud density according to the point cloud horizontal position coordinate difference, the point cloud longitudinal position coordinate difference and the point cloud quantity, and is specifically configured to:

[0199] The point cloud density is obtained through the following formula:

[0200]

[0201] wherein, ρ is the point cloud density, α x is the point cloud horizontal position coordinate difference, α y is the point cloud longitudinal position coordinate difference.

[0202] In one embodiment, the HRRP of the road point cloud data includes the amplitude values of the target points in the road point cloud data;

[0203] The HRRP geometric features include at least one of an equivalent scattering center number, an equivalent target size, an HRRP signal entropy, an HRRP signal standard deviation, an HRRP signal deviation, an HRRP signal irregularity, an HRRP signal echo energy, an HRRP signal energy ratio, an HRRP signal skewness, and an HRRP signal kurtosis.

[0204] The processor is further configured to:

[0205] The equivalent scattering center number is obtained by the following formula:

[0206]

[0207] Num sca is the equivalent scattering center number, y(i) is an amplitude of a target point i, m is an average of amplitudes of target points in the road point cloud data, N is a total number of target points in the road point cloud data, when y(i)-m≥0, ε(y(i)-m)=1, and when y(i)-m<0, ε(y(i)-m)=0;

[0208] The equivalent target size is obtained by the following formula:

[0209] The equivalent target size is obtained according to position coordinates and amplitudes of a target point with the largest amplitude and a target point with the smallest amplitude in the HRRP;

[0210] The HRRP signal entropy is obtained by the following formula:

[0211]

[0212] Entropy is the HRRP signal entropy;

[0213] The HRRP signal standard deviation is obtained by the following formula:

[0214]

[0215] Std is the HRRP signal standard deviation;

[0216] The HRRP signal deviation is obtained by the following formula:

[0217]

[0218] Deviation is the HRRP signal deviation;

[0219] The HRRP signal irregularity is obtained by the following formula:

[0220]

[0221] wherein Irr is the irregularity of the HRRP signal, y(i-1) is the amplitude of the target point i-1 before the target point i, y(i+1) is the amplitude of the target point i+1 after the target point i;

[0222] The echo energy of the HRRP signal is obtained by the following formula:

[0223]

[0224] wherein TP is the echo energy of the HRRP signal;

[0225] The energy ratio of the HRRP signal is obtained by the following way:

[0226] The amplitudes of the target points in the HRRP are sorted in descending order to obtain a sorted HRRP, and the sum of the squares of the amplitudes of the first specified number of target points in the sorted HRRP is divided by the total sum of the squares of the amplitudes of the target points in the HRRP to obtain the energy ratio of the HRRP signal, wherein the specified number is equal to the average value of the amplitudes of the target points;

[0227] The skewness of the HRRP signal is obtained by the following formula:

[0228]

[0229] wherein Skewness is the skewness of the HRRP signal, and μ is the average value of the absolute values of the amplitudes of the target points;

[0230] The kurtosis of the HRRP signal is obtained by the following formula:

[0231]

[0232] wherein kurtosis is the kurtosis of the HRRP signal.

[0233] In one embodiment, the processor is further configured to:

[0234] The HRRP power spectrum feature of the road point cloud data is obtained by the following way:

[0235] For the amplitude of any target point in the HRRP of the road point cloud data, the amplitude of the target point is subjected to Fourier transform to obtain the frequency domain amplitude of the target point; and

[0236] The square of the frequency domain amplitude of the target point is determined as the power spectrum of the target point;

[0237] The power spectrum of each target point in the HRRP of the road point cloud data is determined as a HRRP power spectrum feature of the road point cloud data.

[0238] According to a third aspect provided by the embodiments of the present disclosure, a computer storage medium is provided, which stores a computer program for executing the method according to the first aspect. BRIEF DESCRIPTION OF DRAWINGS

[0239] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present disclosure, and other drawings can be obtained by those skilled in the art without creative labor.

[0240] Figure 1 One of the application scenario schematic diagrams according to an embodiment of the present disclosure;

[0241] Figure 2 Another application scenario schematic diagram according to an embodiment of the present disclosure;

[0242] Figure 3 One of the flow schematic diagrams of the multi-target recognition method in a complex traffic environment according to an embodiment of the present disclosure;

[0243] Figure 4 A road point cloud data schematic diagram according to an embodiment of the present disclosure;

[0244] Figure 5 A flow schematic diagram of the method for determining the road point cloud data according to an embodiment of the present disclosure;

[0245] Figure 6 A road image schematic diagram according to an embodiment of the present disclosure;

[0246] Figure 7 A flow schematic diagram of the training of the target classification model according to an embodiment of the present disclosure;

[0247] Figure 8 A schematic diagram of the SVM model based on the decision tree according to an embodiment of the present disclosure;

[0248] Figure 9 A multi-target recognition device in a complex traffic environment according to an embodiment of the present disclosure;

[0249] Figure 10 A structural schematic diagram of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0250] To make the purposes, technical solutions, and advantages of the embodiments of the present disclosure clearer, the technical solutions in the embodiments of the present disclosure will be described clearly and completely below with reference to the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are some but not all of the embodiments of the present disclosure. Based on the embodiments in the present disclosure, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present disclosure.

[0251] The term “and / or” in the embodiments of the present disclosure describes the association relationship of the associated objects, and indicates that there can be three relationships, for example, A and / or B can represent the three cases of A existing alone, A and B existing simultaneously, and B existing alone. The character “ / ” generally represents an “or” relationship between the associated objects before and after it.

[0252] The application scenarios described in the embodiments of the present disclosure are used to more clearly illustrate the technical solutions of the embodiments of the present disclosure, and do not constitute a limitation on the technical solutions provided by the embodiments of the present disclosure. A person of ordinary skill in the art can know that, as new application scenarios appear, the technical solutions provided by the embodiments of the present disclosure are also applicable to similar technical problems. In the description of the present disclosure, unless otherwise specified, “multiple” means two or more.

[0253] In the prior art, the manner of realizing traffic road vehicle recognition is based on the classification result obtained from the point cloud data generated by single detection of a target. However, in an actual road target classification task, the classification effect of this method is easily affected by abnormal detection results, and the accuracy of target recognition relying on only single-frame point cloud data in a crowded and complex traffic environment is low.

[0254] Therefore, the present disclosure provides a multi-target recognition method in a complex traffic environment. The point cloud feature set of single-frame road point cloud data is determined by first extracting the point cloud feature, HRRP geometric feature, and HRRP power spectrum feature of the single-frame road point cloud data, and then the point cloud feature sequence is obtained by jointly processing the point cloud feature sets of multiple frames of road point cloud data. Target recognition is performed based on the point cloud feature sequence to obtain the categories of each target object. Thus, in the embodiments of the application, multiple features in single-frame road point cloud data are extracted to accurately obtain the point cloud feature set of each type of target object, and the point cloud feature set of multiple frames of road point cloud data in a time sequence is fused to obtain a high-dimensional point cloud feature sequence, so that more detailed information of each target object in a complex traffic environment can be captured to improve the accuracy of multi-target recognition in a complex traffic environment. The scheme of the present disclosure will be described in detail below with reference to the drawings.

[0255] As Figure 1The diagram illustrates an application scenario for a multi-target recognition method in a complex traffic environment. This scenario includes a millimeter-wave radar 110, a server 120, and a terminal device 130. Figure 1 As shown, this application scenario uses an electronic device as a server as an example. However, this application embodiment does not limit the electronic device; the electronic device can be a server or a terminal device, and can be configured according to actual conditions. Server 120 can be implemented using a single server or multiple servers. Server 120 can be implemented using a physical server or a virtual server.

[0256] In one possible application scenario, the millimeter-wave radar 110 performs real-time detection of the target road, obtaining road point cloud data. The server 120 acquires a specified number of frames of road point cloud data at specified intervals. Then, for any given frame of road point cloud data, the server 120 projects the road point cloud data onto the radial distance of the millimeter-wave radar to obtain the high-resolution range signal (HRRP) of the road point cloud data, and extracts point cloud features from the road point cloud data to obtain the point cloud features of the road point cloud data. Based on the point cloud features and HRRP geometric characteristics of the road point cloud data... The system extracts HRRP geometric features and HRRP power spectrum features to obtain a point cloud feature set of the road point cloud data. The HRRP geometric features and HRRP power spectrum features are obtained by feature extraction of the HRRP of the road point cloud data. The server 120 obtains a point cloud feature sequence based on the point cloud feature set of the road point cloud data for a specified number of frames. The point cloud feature sequence is input into a pre-trained target classification model to obtain the category of each target object in the road point cloud data for the specified number of frames. The category of each target object in the road point cloud data for the specified number of frames is then sent to the terminal device 130 for display.

[0257] like Figure 2As shown, it is another application scenario in the embodiment of the present application, which includes a millimeter wave radar 110 and a terminal device 130. The millimeter wave radar 110 detects a target road in real time to obtain road point cloud data of the target road. The terminal device 130 obtains road point cloud data of a specified number of frames every specified time interval. Then, the terminal device 130 projects the road point cloud data on the radial distance of the millimeter wave radar to obtain a high-resolution range profile (HRRP) of the road point cloud data, and extracts point cloud features of the road point cloud data to obtain point cloud features of the road point cloud data. According to the point cloud features, HRRP geometric features and HRRP power spectrum features of the road point cloud data, a point cloud feature set of the road point cloud data is obtained, wherein the HRRP geometric features and the HRRP power spectrum features are obtained by extracting features of the HRRP of the road point cloud data. The terminal device 130 obtains a point cloud feature sequence according to the point cloud feature set of the road point cloud data of the specified number of frames. The point cloud feature sequence is input into a pre-trained target classification model to obtain the category of each target object in the road point cloud data of the specified number of frames, and the category of each target object in the road point cloud data of the specified number of frames is displayed.

[0258] wherein, Figure 1 The server 120 and the terminal device 130 can interact information through a communication network, wherein the communication network can adopt a wireless communication mode or a wired communication mode.

[0259] For example, the server 120 can access the network through a cellular mobile communication technology to communicate with the terminal device 130, wherein the cellular mobile communication technology includes, for example, a 5th Generation Mobile Networks (5G) technology.

[0260] Optionally, the server 120 can access the network through a short-range wireless communication mode to communicate with the terminal device 130, wherein the short-range wireless communication mode includes, for example, a Wireless Fidelity (Wi-Fi) technology.

[0261] In the description of the present application, only a single millimeter wave radar 110, a single server 120 and a single terminal device 130 are described in detail, but those skilled in the art should understand that the millimeter wave radar 110, the server 120 and the terminal device 130 shown are intended to represent the operation of the millimeter wave radar 110, the server 120 and the terminal device 130 involved in the technical solution of the present application. It is not implied that there is a limitation on the number, type or location of the millimeter wave radar 110, the server 120 and the terminal device 130. It should be noted that if additional modules are added to the illustrated environment or individual modules are removed therefrom, the underlying concept of the example embodiments of the present application will not change.

[0262] It should be noted that the multi-target recognition method in a complex traffic environment proposed in the present application is not only applicable to Figure 1 and Figure 2 the application scenarios shown, but also applicable to any multi-target recognition device in a complex traffic environment.

[0263] The multi-target recognition method in a complex traffic environment of the example embodiments of the present application will be described below in conjunction with the above-described application scenarios and with reference to the accompanying drawings. It should be noted that the above-described application scenarios are only shown to facilitate understanding of the method and principles of the present application, and the embodiments of the present application are not limited in this respect.

[0264] As shown in Figure 3 , it is a flowchart of the multi-target recognition method in a complex traffic environment of the present disclosure, which can include the following steps:

[0265] Step 301: Obtain road point cloud data of a specified number of frames every specified time interval, wherein the road point cloud data is obtained by detecting the target road using a millimeter wave radar.

[0266] As shown in Figure 4 , it is a schematic diagram of road point cloud data. Since only vehicles exist in the target road, each target point in the road point cloud data is a target point of a vehicle.

[0267] In this embodiment, the road point cloud data is obtained by performing a two-dimensional FFT (Fast Fourier transform) on the original signal sampling data containing target object activity information acquired by millimeter-wave radar to obtain two-dimensional map data containing the relative speed and relative distance information of the target object; then, peak detection is performed on the two-dimensional map data using the CFAR (Constant False Alarm Rate) algorithm to obtain peak data containing the speed and distance of the target object; then, DOA (Direction of Arrival) angle acquisition is performed on the extracted peak data to obtain point cloud data containing relative distance, heading angle, and pitch angle; finally, the spatial coordinate information of the point cloud data is obtained through trigonometric function theorems, and the motion feature set is obtained through calculation to obtain the road point cloud data.

[0268] It should be noted that the method of obtaining road point cloud data from the raw signals acquired by millimeter-wave radar in this embodiment can be set according to the actual situation. The method described above is only for illustrative purposes and does not limit the method of obtaining road point cloud data. Furthermore, the target road, specified duration, and specified number of frames in this embodiment can be set according to the actual situation, and this embodiment does not limit the specified duration and specified number of frames.

[0269] Step 302: For any frame of road point cloud data, project the road point cloud data onto the radial distance of the millimeter-wave radar to obtain the high-resolution range signal HRRP of the road point cloud data, and extract point cloud features from the road point cloud data to obtain the point cloud features of the road point cloud data.

[0270] In this embodiment of the application, the road point cloud data is obtained by projection as follows: if the road point cloud data of the b-th frame is a set P = {p1,...,p...} n}, where P i =(d i ,m i ), d i m represents the spatial coordinates of target point i. i The state vector of target point i is represented, which includes the radar cross section and signal-to-noise ratio of target point i. The spatial location coordinates of the road point cloud data are converted into a three-dimensional map of RDA (Radar Data Acquisition) and then inverse Fourier transform is performed on it in the angle and velocity dimensions to obtain the HRRP of the road point cloud data of the b-th frame.

[0271] The projection mode described in the foregoing of the embodiments of the present application is only used for illustration and does not limit the way of obtaining HRRP, and the specific way can be set according to actual conditions.

[0272] wherein the road point cloud data comprises position coordinates, radial velocity and echo intensity of each target point; and any one point cloud feature comprises at least one of point cloud quantity, horizontal position coordinate of point cloud centroid, vertical position coordinate of point cloud centroid, ratio of horizontal position coordinate to vertical position coordinate of point cloud centroid, relative radar distance of point cloud centroid, radial velocity of point cloud centroid, radial velocity variance of point cloud centroid, radial velocity change difference of point cloud centroid, minimum value of point cloud echo intensity, maximum value of point cloud echo intensity, mean value of point cloud echo intensity, point cloud echo intensity variance, point cloud horizontal position coordinate difference, point cloud vertical position coordinate difference, point cloud horizontal position coordinate variance, point cloud vertical position coordinate variance and point cloud density; the determination method of each point cloud feature is described as follows:

[0273] 1. Point cloud quantity: the total quantity of each target point in the road point cloud data is determined as the point cloud quantity.

[0274] 2. Horizontal position coordinate of point cloud centroid: the mean value of the horizontal position coordinates of each target point in the road point cloud data is determined as the horizontal position coordinate of the point cloud centroid. Wherein, the horizontal position coordinate of the point cloud centroid can be obtained by formula (1):

[0275]

[0276] wherein, is the horizontal position coordinate of the point cloud centroid, x i is the horizontal position coordinate of target point i in the road point cloud data, i∈[1, N], N is the total quantity of each target point in the road point cloud data.

[0277] 3. Vertical position coordinate of point cloud centroid: the mean value of the vertical position coordinates of each target point in the road point cloud data is determined as the vertical position coordinate of the point cloud centroid. Wherein, the vertical position coordinate of the point cloud centroid can be obtained by formula (2):

[0278]

[0279] wherein, is the vertical position coordinate of the point cloud centroid, y i is the vertical position coordinate of target point i in the road point cloud data.

[0280] 4. Ratio of horizontal position coordinate to vertical position coordinate of point cloud centroid: the horizontal position coordinate of the point cloud centroid is divided by the vertical position coordinate of the point cloud centroid to obtain the ratio of horizontal position coordinate to vertical position coordinate of the point cloud centroid. Wherein, the ratio of horizontal position coordinate to vertical position coordinate of the point cloud centroid can be obtained by formula (3):

[0281]

[0282] wherein σ is the ratio of the horizontal coordinate of the point cloud centroid to the vertical coordinate of the point cloud centroid.

[0283] 5. Point cloud centroid relative radar distance: the point cloud centroid relative radar distance is obtained according to the horizontal position coordinate of the point cloud centroid and the vertical position coordinate of the point cloud centroid. Wherein the point cloud centroid relative radar distance can be obtained by formula (4):

[0284]

[0285] wherein R is the point cloud centroid relative radar distance.

[0286] 6. Point cloud centroid radial velocity: the average value of the radial velocity of each target point in the road point cloud data is determined as the point cloud centroid radial velocity. Wherein the point cloud centroid radial velocity can be obtained by formula (5):

[0287]

[0288] wherein, is the point cloud centroid radial velocity, v i is the radial velocity of target point i in the road point cloud data.

[0289] 7. Point cloud centroid radial velocity variance: the point cloud centroid radial velocity variance is obtained according to the radial velocity of each target point in the road point cloud data and the point cloud centroid radial velocity. Wherein the point cloud centroid radial velocity variance can be obtained by formula (6):

[0290]

[0291] wherein δ v is the point cloud centroid radial velocity variance, v i is the radial velocity of target point i in the road point cloud data, is the point cloud centroid radial velocity.

[0292] 8. Point cloud centroid radial velocity change difference: the difference between the radial velocity of the target point with the largest value in the road point cloud data and the radial velocity of the target point with the smallest value is determined as the point cloud centroid radial velocity change difference. Wherein the point cloud centroid radial velocity change difference can be obtained by formula (7):

[0293] a v = v max - v min ……(7);

[0294] wherein a vis the radial velocity of the target point with the maximum value in the road point cloud data. max is the radial velocity of the target point with the maximum value in the road point cloud data. min is the radial velocity of the target point with the minimum value in the road point cloud data.

[0295] 9. Point cloud echo intensity minimum value: the echo intensity of the target point with the minimum value in the road point cloud data is determined as the point cloud echo intensity minimum value. The point cloud echo intensity minimum value can be obtained by formula (8):

[0296] dB min = min(dB i ) …… (8).

[0297] wherein, dB min is the point cloud echo intensity minimum value, and dB i is the echo intensity of the target point i in the road point cloud data, i ∈ [1, N].

[0298] 10. Point cloud echo intensity maximum value: the echo intensity of the target point with the maximum value in the road point cloud data is determined as the point cloud echo intensity maximum value. The point cloud echo intensity maximum value can be obtained by formula (9):

[0299] dB max = max(dB i ) …… (9).

[0300] wherein, dB max is the point cloud echo intensity maximum value.

[0301] 11. Point cloud echo intensity mean value: the mean value of the echo intensities of all target points in the road point cloud data is determined as the point cloud echo intensity mean value. The point cloud echo intensity mean value can be obtained by formula (10):

[0302]

[0303] wherein, is the point cloud echo intensity mean value, and dB i is the echo intensity of the target point i.

[0304] 12. Point cloud echo intensity variance: the point cloud echo intensity variance is obtained according to the echo intensities of all target points in the road point cloud data and the point cloud echo intensity mean value. The point cloud echo intensity variance can be obtained by formula (11):

[0305]

[0306] wherein, δ dBPoint cloud echo intensity variance, dB i Echo intensity of target point i in road point cloud data, Point cloud echo intensity mean 13. Point cloud lateral position coordinate difference: subtract the lateral position coordinate of the target point with the largest value from the lateral position coordinate of the target point with the smallest value in the road point cloud data to obtain the point cloud lateral position coordinate difference.

[0307] 13. Point cloud lateral position coordinate difference: subtract the lateral position coordinate of the target point with the largest value from the lateral position coordinate of the target point with the smallest value in the road point cloud data to obtain the point cloud lateral position coordinate difference. Wherein, the point cloud lateral position coordinate difference can be obtained by formula (12):

[0308] a x = x max - x min … (12);

[0309] Wherein, a x is the point cloud lateral position coordinate difference, x max is the lateral position coordinate of the target point with the largest value in the road point cloud data, x min is the lateral position coordinate of the target point with the largest value in the road point cloud data.

[0310] 14. Point cloud longitudinal position coordinate difference: subtract the longitudinal position coordinate of the target point with the largest value from the longitudinal position coordinate of the target point with the smallest value in the road point cloud data to obtain the point cloud longitudinal position coordinate difference. Wherein, the point cloud longitudinal position coordinate difference can be obtained by formula (13):

[0311] a y = y max - y min … (13);

[0312] Wherein, a y is the point cloud longitudinal position coordinate difference, y max is the longitudinal position coordinate of the target point with the largest value in the road point cloud data, y min is the longitudinal position coordinate of the target point with the largest value in the road point cloud data.

[0313] 15. Point cloud lateral position coordinate variance: according to the lateral position coordinates of each target point in the road point cloud data and the lateral position coordinate of the point cloud centroid, the point cloud lateral position coordinate variance is obtained. Wherein, the point cloud lateral position coordinate variance can be obtained by formula (14):

[0314]

[0315] Wherein, δ x is the point cloud lateral position coordinate variance, xi is a lateral position coordinate of the target point i in the road point cloud data, is a lateral position coordinate of the point cloud centroid.

[0316] 16. Point cloud longitudinal position coordinate variance: the point cloud longitudinal position coordinate variance is obtained according to the longitudinal position coordinates of each target point in the road point cloud data and the longitudinal position coordinate of the point cloud centroid. The point cloud longitudinal position coordinate variance can be obtained by formula (15):

[0317]

[0318] wherein, δ y is the point cloud longitudinal position coordinate variance, y i is a longitudinal position coordinate of the target point i in the road point cloud data, is a longitudinal position coordinate of the point cloud centroid.

[0319] 17. Point cloud density: the point cloud density is obtained according to the point cloud lateral position coordinate difference, the point cloud longitudinal position coordinate difference and the point cloud quantity. The point cloud density can be obtained by formula (16):

[0320]

[0321] wherein, ρ is the point cloud density, α x is the point cloud lateral position coordinate difference, α y is the point cloud longitudinal position coordinate difference.

[0322] Step 303: obtaining the point cloud feature set of each road point cloud data according to the point cloud features, HRRP geometric features and HRRP power spectrum features of the road point cloud data, wherein the HRRP geometric features and the HRRP power spectrum features are obtained by feature extraction on the HRRP of the road point cloud data;

[0323] wherein, the HRRP of the road point cloud data includes the amplitude of each target point in the road point cloud data; the HRRP geometric features include at least one of the equivalent scattering center quantity, the equivalent target size, the HRRP signal entropy, the HRRP signal standard deviation, the HRRP signal deviation, the HRRP signal irregularity, the HRRP signal echo energy, the HRRP signal energy ratio, the HRRP signal skewness and the HRRP signal kurtosis;

[0324] Next, the way of extracting the HRRP geometric features in the embodiments of the present application is introduced:

[0325] 1. Equivalent scattering center quantity: the equivalent scattering center quantity can be obtained by formula (17):

[0326]

[0327] wherein, Num sca is the number of equivalent scattering centers, y(i) is the amplitude of target point i, m is the average of the amplitudes of the target points in the road point cloud data, N is the total number of target points in the road point cloud data, when y(i)-m≥0, ε(y(i)-m)=1, when y(i)-m<0, ε(y(i)-m)=0.

[0328] 2. Equivalent target size: according to the position coordinates of the target point with the largest amplitude in the HRRP and the position coordinates of the target point with the smallest amplitude, the equivalent target size is obtained.

[0329] In one embodiment, the transverse position coordinates of the target point with the largest amplitude in the HRRP are subtracted from the transverse position coordinates of the target point with the smallest amplitude to obtain the length of the equivalent target, and the longitudinal position coordinates of the target point with the largest amplitude in the HRRP are subtracted from the longitudinal position coordinates of the target point with the smallest amplitude to obtain the width of the equivalent target, and the length of the equivalent target and the width of the equivalent target are determined as the equivalent target size.

[0330] 3. HRRP signal entropy: the HRRP signal entropy is obtained by formula (18):

[0331]

[0332] wherein, Entropy is the HRRP signal entropy.

[0333] 4. HRRP signal standard deviation: the HRRP signal standard deviation is obtained by formula (19):

[0334]

[0335] wherein, Std is the HRRP signal standard deviation.

[0336] 5. HRRP signal deviation: the HRRP signal deviation is obtained by formula (20):

[0337]

[0338] Deviation is the HRRP signal deviation.

[0339] 6. HRRP signal irregularity: the HRRP signal irregularity is obtained by formula (21):

[0340]

[0341] wherein, Irr is the irregularity of HRRP signal, y(i-1) is the amplitude of the target point i-1 before the target point i, y(i+1) is the amplitude of the target point i+1 after the target point i.

[0342] 7. HRRP signal echo energy: the HRRP signal echo energy is obtained by formula (22):

[0343]

[0344] wherein, TP is the HRRP signal echo energy.

[0345] 8. HRRP signal energy ratio: the amplitudes of each target point in the HRRP are sorted in descending order to obtain a sorted HRRP; and the sum of the squares of the amplitudes of the first specified number of each target point in the sorted HRRP is divided by the total sum of the squares of the amplitudes of each target point in the HRRP to obtain the HRRP signal energy ratio, wherein the specified number is equal to the average value of the amplitudes of each target point. The HRRP signal energy ratio is obtained by formula (23):

[0346]

[0347] wherein, P m is the HRRP signal energy ratio, x(i) is the amplitude of the target point i in the sorted HRRP.

[0348] 9. HRRP signal skewness: the HRRP signal skewness is obtained by formula (24):

[0349]

[0350] wherein, Skewness is the HRRP signal skewness, μ is the average value of the absolute values of the amplitudes of each target point. The average value of the absolute values of the amplitudes of each target point can be obtained by formula (25):

[0351]

[0352] wherein, μ is the average value of the absolute values of the amplitudes of each target point.

[0353] 10. HRRP signal kurtosis: the HRRP signal kurtosis is obtained by formula (26):

[0354]

[0355] wherein, kurtosis is the kurtosis of the HRRP signal.

[0356] Next, the way of determining the HRRP power spectrum features of the road point cloud data is introduced, as shown in Figure 5 The flowchart for determining the road point cloud data includes the following steps:

[0357] Step 501: For the amplitude of any target point in the HRRP of the road point cloud data, Fourier transform is performed on the amplitude of the target point to obtain the frequency domain amplitude of the target point.

[0358] Step 502: The square of the frequency domain amplitude of the target point is determined as the power spectrum of the target point; wherein the power spectrum of the target point can be obtained by formula (27):

[0359] D(i)=w(i) 2 ……(27);

[0360] Wherein, D(i) is the power spectrum of the target point i, and w(i) is the frequency domain amplitude of the target point i.

[0361] Step 503: The power spectrum of each target point in the HRRP of the road point cloud data is determined as the HRRP power spectrum features of the road point cloud data.

[0362] Step 304: Obtain the point cloud feature sequence according to the point cloud feature set of the road point cloud data of the specified frame number.

[0363] In an embodiment, the point cloud feature sequence can be obtained in the following two ways:

[0364] Way one: the point cloud feature set of the road point cloud data of the specified frame number is composed into a feature sequence, and the feature sequence is determined as the point cloud feature sequence.

[0365] Way two: the point cloud feature set of the road point cloud data of the specified frame number is composed into a feature sequence, for any feature in the feature sequence, the number of the feature in the feature sequence is counted, the number of the feature in the feature sequence is divided by the total number of features in the feature sequence to obtain the probability density of the feature, the features with probability density less than the preset threshold in the feature sequence are deleted from the feature sequence to obtain the point cloud feature sequence.

[0366] Step 305: input the point cloud feature sequence into the pre-trained target classification model to obtain the category of each target object in the road point cloud data of the specified frame number.

[0367] The target object in the embodiment of the application can be a motor vehicle, a pedestrian, and a non-motor vehicle, etc., which can be set according to actual conditions, and the embodiment of the application does not limit here.

[0368] As shown in Figure 6 , a road image corresponding to any frame of road point cloud data, from which it can be seen that the positions of each target object in the road image and the categories corresponding to each target object.

[0369] Next, the training method of the target classification model in the embodiments of the present application is briefly introduced, as shown in Figure 7 , a schematic diagram of the training process of the target classification model, which can include the following steps:

[0370] Step 701: obtaining a training sample, wherein the training sample includes a point cloud feature sequence corresponding to multiple frames of road point cloud data and standard positions and labeled categories of each target object in the multiple frames of road point cloud data, and any one road image includes a labeled position of a target object and a labeled category of the target object;

[0371] Step 702: inputting the training sample into the target classification model for detection to obtain predicted positions and predicted categories of each target object in the multiple frames of road point cloud data in the training sample;

[0372] Step 703: obtaining a first loss value based on the predicted positions and the labeled positions of each target object in the multiple frames of road point cloud data;

[0373] Step 704: obtaining a second loss value based on the predicted categories and the labeled categories of each target object in the multiple frames of road point cloud data;

[0374] Step 705: adding the first loss value and the second loss value to obtain a total loss value;

[0375] Step 706: determining whether the total loss value is less than a specified loss value, if yes, executing step 707, and if no, executing step 708;

[0376] Step 707: adjusting the model parameters of the decision tree-based SVM model and returning to execute step 702;

[0377] Step 708: ending the training of the target classification model to obtain the trained target classification model.

[0378] The target classification model in the embodiments of the present application uses a decision tree-based SVM (Support Vector Machine) model. The decision tree in the embodiments of the present application adopts a non-balanced binary tree architecture, as shown in Figure 8 , the SVM1 of the root node divides three categories of targets into two categories: one category of motor vehicles, and one category of pedestrians and non-motor vehicles; the sub-node SVM2 of the second layer further distinguishes pedestrians and non-motor vehicles.

[0379] It should be noted that the target classification model described above, which is an SVM model based on decision trees, is only used as an example and is not intended to limit the target classification model in the embodiments of this application. The target classification model in the embodiments of this application can be set according to the actual situation.

[0380] Based on the same disclosed concept, the multi-target recognition method in complex traffic environments described above can also be implemented by a multi-target recognition device in complex traffic environments. The effect of this multi-target recognition device in complex traffic environments is similar to that of the aforementioned method, and will not be repeated here.

[0381] Figure 9 This is a schematic diagram of the structure of a multi-target recognition device in a complex traffic environment according to an embodiment of the present disclosure.

[0382] like Figure 9 As shown, the multi-target recognition device 900 in complex traffic environments disclosed herein may include a road point cloud data acquisition module 910, a point cloud feature extraction module 920, a point cloud feature set determination module 930, a point cloud feature sequence determination module 940, and a target object category determination module 950.

[0383] The road point cloud data acquisition module 910 is used to acquire a specified number of road point cloud data at specified intervals, wherein the road point cloud data is obtained by detecting the target road using millimeter-wave radar.

[0384] The point cloud feature extraction module 920 is used to project any frame of road point cloud data onto the radial distance of the millimeter-wave radar to obtain the high-resolution range signal HRRP of the road point cloud data, and to extract point cloud features from the road point cloud data to obtain the point cloud features of the road point cloud data; and,

[0385] The point cloud feature set determination module 930 is used to obtain the point cloud feature set of each road point cloud data according to the point cloud features, HRRP geometric features and HRRP power spectrum features of the road point cloud data, wherein the HRRP geometric features and the HRRP power spectrum features are obtained by feature extraction of the HRRP of the road point cloud data.

[0386] The point cloud feature sequence determination module 940 is used to obtain a point cloud feature sequence based on the point cloud feature set of the road point cloud data of the specified number of frames;

[0387] The target object category determination module 950 is used to input the point cloud feature sequence into a pre-trained target classification model to obtain the category of each target object in the road point cloud data of the specified number of frames.

[0388] In one embodiment, the road point cloud data comprises position coordinates, radial velocities and echo intensities of each target point; and any one point cloud feature comprises at least one of point cloud quantity, horizontal position coordinate of point cloud centroid, vertical position coordinate of point cloud centroid, ratio of horizontal and vertical position coordinates of point cloud centroid, relative radar distance of point cloud centroid, radial velocity of point cloud centroid, radial velocity variance of point cloud centroid, radial velocity variation difference of point cloud centroid, minimum value of point cloud echo intensity, maximum value of point cloud echo intensity, mean value of point cloud echo intensity, variance of point cloud echo intensity, horizontal position coordinate difference of point cloud, vertical position coordinate difference of point cloud, horizontal position coordinate variance of point cloud, vertical position coordinate variance of point cloud and point cloud density.

[0389] The point cloud feature extraction module 920 is specifically configured to:

[0390] determine the total number of target points in the road point cloud data as the point cloud quantity; and / or,

[0391] determine the mean value of the horizontal position coordinates of the target points in the road point cloud data as the horizontal position coordinate of the point cloud centroid; and / or,

[0392] determine the mean value of the vertical position coordinates of the target points in the road point cloud data as the vertical position coordinate of the point cloud centroid; and / or,

[0393] divide the horizontal position coordinate of the point cloud centroid by the vertical position coordinate of the point cloud centroid to obtain the ratio of horizontal and vertical position coordinates of the point cloud centroid; and / or,

[0394] obtain the relative radar distance of the point cloud centroid according to the horizontal position coordinate of the point cloud centroid and the vertical position coordinate of the point cloud centroid; and / or,

[0395] determine the mean value of the radial velocities of the target points in the road point cloud data as the radial velocity of the point cloud centroid; and / or,

[0396] obtain the radial velocity variance of the point cloud centroid according to the radial velocities of the target points in the road point cloud data and the radial velocity of the point cloud centroid; and / or,

[0397] determine the difference between the radial velocity of the target point with the maximum value and the radial velocity of the target point with the minimum value in the road point cloud data as the radial velocity variation difference of the point cloud centroid; and / or,

[0398] determine the echo intensity of the target point with the minimum value in the road point cloud data as the minimum value of the point cloud echo intensity; and / or,

[0399] determine the echo intensity of the target point with the maximum value in the road point cloud data as the maximum value of the point cloud echo intensity; and / or,

[0400] determining a mean value of echo intensity of each target point in the road point cloud data as the point cloud echo intensity mean value;

[0401] obtaining a point cloud echo intensity variance according to the echo intensity of each target point in the road point cloud data and the point cloud echo intensity mean value;

[0402] subtracting the lateral position coordinate of the target point with the maximum value from the lateral position coordinate of the target point with the minimum value in the road point cloud data to obtain the point cloud lateral position coordinate difference;

[0403] subtracting the longitudinal position coordinate of the target point with the minimum value from the longitudinal position coordinate of the target point with the maximum value in the road point cloud data to obtain the point cloud longitudinal position coordinate difference;

[0404] obtaining the point cloud lateral position coordinate variance according to the lateral position coordinate of each target point in the road point cloud data and the lateral position coordinate of the point cloud centroid;

[0405] obtaining the point cloud longitudinal position coordinate variance according to the longitudinal position coordinate of each target point in the road point cloud data and the longitudinal position coordinate of the point cloud centroid;

[0406] obtaining the point cloud density according to the point cloud lateral position coordinate difference, the point cloud longitudinal position coordinate difference and the point cloud quantity.

[0407] In one embodiment, the point cloud feature extraction module 920 performs the determination of the mean value of the lateral position coordinate of each target point in the road point cloud data as the lateral position coordinate of the point cloud centroid, specifically for:

[0408] the lateral position coordinate of the point cloud centroid is obtained by the following formula:

[0409]

[0410] wherein, is the lateral position coordinate of the point cloud centroid, x i is the lateral position coordinate of the target point i in the road point cloud data, i ∈ [1, N], N is the total quantity of each target point in the road point cloud data;

[0411] The point cloud feature extraction module 920 performs the determination of the mean value of the longitudinal position coordinate of each target point in the road point cloud data as the longitudinal position coordinate of the point cloud centroid, specifically for:

[0412] the longitudinal position coordinate of the point cloud centroid is obtained by the following formula:

[0413]

[0414] wherein, is a longitudinal position coordinate of the point cloud centroid, y i is a longitudinal position coordinate of target point i in the road point cloud data;

[0415] The point cloud feature extraction module 920 performs the division of the lateral position coordinate of the point cloud centroid by the longitudinal position coordinate of the point cloud centroid to obtain the point cloud centroid lateral-longitudinal coordinate ratio, specifically for:

[0416] The point cloud centroid lateral-longitudinal coordinate ratio is obtained by the following formula:

[0417]

[0418] wherein σ is the point cloud centroid lateral-longitudinal coordinate ratio;

[0419] The point cloud feature extraction module 920 performs the obtaining of the point cloud centroid relative radar distance according to the lateral position coordinate of the point cloud centroid and the longitudinal position coordinate of the point cloud centroid, specifically for:

[0420] The point cloud centroid relative radar distance is obtained by the following formula:

[0421]

[0422] wherein R is the point cloud centroid relative radar distance;

[0423] The point cloud feature extraction module 920 performs the determination of the point cloud centroid radial velocity as the average value of the radial velocities of each target point in the road point cloud data, specifically for:

[0424] The point cloud centroid radial velocity is obtained by the following way:

[0425]

[0426] wherein, is the point cloud centroid radial velocity, v i is a radial velocity of target point i in the road point cloud data;

[0427] The point cloud feature extraction module 920 performs the obtaining of the point cloud centroid radial velocity variance according to the radial velocities of each target point in the road point cloud data and the point cloud centroid radial velocity, specifically for:

[0428] The point cloud centroid radial velocity variance is obtained by the following formula:

[0429]

[0430] wherein δ v is the point cloud centroid radial velocity variance, vi a radial velocity of a target point i in the road point cloud data, a point cloud centroid radial velocity;

[0431] The point cloud feature extraction module 920 performs the determination of the difference between the radial velocity of the target point with the maximum value in the road point cloud data and the radial velocity of the target point with the minimum value as the point cloud centroid radial velocity change difference, and specifically for:

[0432] The point cloud centroid radial velocity change difference is obtained by the following formula:

[0433] a v = v max - v min ;

[0434] Wherein, a v is the point cloud centroid radial velocity change difference, v max is the radial velocity of the target point with the maximum value in the road point cloud data, and v min is the radial velocity of the target point with the minimum value in the road point cloud data;

[0435] The point cloud feature extraction module 920 performs the determination of the mean value of the echo intensity of each target point in the road point cloud data as the point cloud echo intensity mean value, and specifically for:

[0436] The point cloud echo intensity mean value is obtained by the following formula:

[0437]

[0438] Wherein, is the point cloud echo intensity mean value, dB i is the echo intensity of the target point i;

[0439] The point cloud feature extraction module 920 performs the determination of the point cloud echo intensity variance according to the echo intensity of each target point in the road point cloud data and the point cloud echo intensity mean value, and specifically for:

[0440] The point cloud echo intensity variance is obtained by the following formula:

[0441]

[0442] Wherein, δ dB is the point cloud echo intensity variance, dB i is the echo intensity of the target point i in the road point cloud data, is the point cloud echo intensity mean value;

[0443] The point cloud feature extraction module 920 performs the operation of subtracting the lateral position coordinate of the target point with the maximum value in the road point cloud data from the lateral position coordinate of the target point with the minimum value to obtain the point cloud lateral position coordinate difference, specifically for:

[0444] a x = x max - x min ;

[0445] wherein a x is the point cloud lateral position coordinate difference, x max is the lateral position coordinate of the target point with the maximum value in the road point cloud data, and x min is the lateral position coordinate of the target point with the maximum value in the road point cloud data;

[0446] The point cloud feature extraction module 920 performs the operation of subtracting the longitudinal position coordinate of the target point with the maximum value in the road point cloud data from the longitudinal position coordinate of the target point with the minimum value to obtain the point cloud longitudinal position coordinate difference, specifically for:

[0447] The point cloud longitudinal position coordinate difference is obtained by the following formula:

[0448] a y = y max - y min ;

[0449] wherein a y is the point cloud longitudinal position coordinate difference, y max is the longitudinal position coordinate of the target point with the maximum value in the road point cloud data, and y min is the longitudinal position coordinate of the target point with the maximum value in the road point cloud data;

[0450] The point cloud feature extraction module 920 performs the operation of obtaining the point cloud lateral position coordinate variance according to the lateral position coordinates of the target points in the road point cloud data and the lateral position coordinate of the point cloud centroid, specifically for:

[0451] The point cloud lateral position coordinate variance is obtained by the following formula:

[0452]

[0453] wherein δ x is the point cloud lateral position coordinate variance, x i is the lateral position coordinate of the target point i in the road point cloud data, is the lateral position coordinate of the point cloud centroid;

[0454] The point cloud feature extraction module 920 performs the obtaining of the point cloud longitudinal position coordinate variance according to the longitudinal position coordinates of each target point in the road point cloud data and the longitudinal position coordinate of the point cloud centroid, and specifically for:

[0455] The point cloud longitudinal position coordinate variance is obtained by the following formula:

[0456]

[0457] wherein δ y is the point cloud longitudinal position coordinate variance, y i is the longitudinal position coordinate of the target point i in the road point cloud data, is the longitudinal position coordinate of the point cloud centroid;

[0458] The point cloud feature extraction module 920 performs the obtaining of the point cloud density according to the point cloud transverse position coordinate difference, the point cloud longitudinal position coordinate difference, and the point cloud quantity, and specifically for:

[0459] The point cloud density is obtained by the following formula:

[0460]

[0461] wherein ρ is the point cloud density, α x is the point cloud transverse position coordinate difference, and α y is the point cloud longitudinal position coordinate difference.

[0462] In one embodiment, the HRRP of the road point cloud data includes the amplitude values of each target point in the road point cloud data; the HRRP geometric features include at least one of the equivalent scattering center quantity, the equivalent target size, the HRRP signal entropy, the HRRP signal standard deviation, the HRRP signal deviation, the HRRP signal irregularity, the HRRP signal echo energy, the HRRP signal energy ratio, the HRRP signal skewness, and the HRRP signal kurtosis; and the device further includes:

[0463] The HRRP geometric feature extraction module 960 is configured to obtain the equivalent scattering center quantity by the following formula:

[0464]

[0465] wherein Num sca is the equivalent scattering center quantity, y(i) is the amplitude value of the target point i, m is the average value of the amplitude values of each target point in the road point cloud data, N is the total quantity of each target point in the road point cloud data, ε(y(i)-m)=1 when y(i)-m≥0, and ε(y(i)-m)=0 when y(i)-m<0;

[0466] The equivalent target size is obtained by the following formula:

[0467] The equivalent target size is obtained according to the position coordinates of the target point with the maximum amplitude and the position coordinates of the target point with the minimum amplitude in the HRRP;

[0468] The HRRP signal entropy is obtained by the following formula:

[0469]

[0470] Wherein, Entropy is the HRRP signal entropy;

[0471] The HRRP signal standard deviation is obtained by the following formula:

[0472]

[0473] Wherein, Std is the HRRP signal standard deviation;

[0474] The HRRP signal deviation is obtained by the following formula:

[0475]

[0476] Deviation is the HRRP signal deviation;

[0477] The HRRP signal irregularity is obtained by the following formula:

[0478]

[0479] Wherein, Irr is the irregularity of the HRRP signal, y(i-1) is the amplitude of the previous target point i-1 of the target point i, and y(i+1) is the amplitude of the next target point i+1 of the target point i;

[0480] The HRRP signal echo energy is obtained by the following formula:

[0481]

[0482] Wherein, TP is the HRRP signal echo energy;

[0483] The HRRP signal energy ratio is obtained by the following formula:

[0484] sort the HRRP in descending order of the amplitude of each target point to obtain a sorted HRRP; and divide the sum of the squares of the amplitudes of a specified number of target points in the sorted HRRP by the sum of the squares of the amplitudes of each target point in the HRRP to obtain the HRRP signal energy ratio, wherein the specified number is equal to the average of the amplitudes of each target point;

[0485] The HRRP signal skewness is obtained by the following formula:

[0486]

[0487] wherein Skewness is the HRRP signal skewness, and μ is the average of the absolute values of the amplitudes of each target point;

[0488] The HRRP signal kurtosis is obtained by the following formula:

[0489]

[0490] wherein urtosis is the HRRP signal kurtosis.

[0491] In one embodiment, the apparatus further comprises:

[0492] The HRRP power spectrum feature extraction module 970 is configured to obtain the HRRP power spectrum features of the road point cloud data by the following manner:

[0493] For the amplitude of any one target point in the HRRP of the road point cloud data, performing Fourier transform on the amplitude of the target point to obtain the frequency domain amplitude of the target point; and

[0494] Determining the square of the frequency domain amplitude of the target point as the power spectrum of the target point;

[0495] Determining the power spectrum of each target point in the HRRP of the road point cloud data as the HRRP power spectrum features of the road point cloud data.

[0496] After introducing the multi-target identification method and apparatus in a complex traffic environment according to an example embodiment of the present disclosure, next, an electronic device according to another example embodiment of the present disclosure is introduced.

[0497] Those skilled in the art can understand that each aspect of the present disclosure can be implemented as a system, a method or a program product. Therefore, each aspect of the present disclosure can be embodied as a whole hardware embodiment, a whole software embodiment (including firmware, microcode, etc.), or an embodiment combining hardware and software aspects, which can be collectively referred to as a "circuit", a "module" or a "system" herein.

[0498] In some possible implementations, the electronic device according to the present disclosure can include at least one processor and at least one computer storage medium. The computer storage medium stores program codes which, when executed by the processor, cause the processor to perform the steps in the multi-target recognition method in a complex traffic environment according to various exemplary embodiments of the present disclosure described above in the specification. For example, the processor can perform steps 301-305 as shown in Figure 3 FIG. 3.

[0499] The electronic device 1000 according to this implementation of the present disclosure will be described below with reference to Figure 10 FIG. 1. Figure 10 The electronic device 1000 shown is merely an example and should not impose any limitation on the function and scope of use of the embodiments of the present disclosure.

[0500] As shown in Figure 10 FIG. 1, the electronic device 1000 is in the form of a general electronic device. The components of the electronic device 1000 can include, but are not limited to, the at least one processor 1001 described above, the at least one computer storage medium 1002 described above, and a bus 1003 connecting different system components, including the computer storage medium 1002 and the processor 1001.

[0501] The bus 1003 represents one or more of several types of bus structures, including a computer storage medium bus or computer storage medium controller, a peripheral bus, a processor bus, or a local bus using any of a variety of bus architectures.

[0502] The computer storage medium 1002 can include readable media in the form of volatile computer storage medium, such as random access computer storage medium (RAM) 1021 and / or cache computer storage medium 1022, and can further include read-only computer storage medium (ROM) 1023.

[0503] The computer storage medium 1002 can further include programs / utilities 1025 with a set of (at least one) program modules 1024, such as an operating system, one or more application programs, other program modules, and program data, each of which or some combination of which can include implementation of a network environment.

[0504] The electronic device 1000 can also communicate with one or more external devices 1004 such as a keyboard or a pointing device, through an input / output (I / O) interface(s) 1005. And, the electronic device 1000 can communicate with one or more networks, such as a local area network (LAN), a wide area network (WAN), and / or the public network, such as the Internet, through a network adapter 1006. As depicted, the network adapter 1006 is in communication with the other components of the electronic device 1000 through the bus 1003. It should be understood that, although not shown, other hardware and / or software components could be used in conjunction with the electronic device 1000. These components, as well as the software components, are meant to encompass existing technology and / or future technology that performs the described functions.

[0505] In some possible embodiments, various aspects of the method for multi-target recognition in complex traffic environment provided by the present disclosure can also be implemented as a program product in the form of a computer-readable medium tangibly embodying a program of instructions executable by a computer device to perform steps of the method for multi-target recognition in complex traffic environment according to various example embodiments of the present disclosure described above in the specification.

[0506] The program product can take any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable storage medium or a computer-readable signal medium. The computer-readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the above. More specific examples (a non-exhaustive list) of the computer-readable storage medium include an electrical connection having one or more wires, a portable disc, a hard disk, a random access computer memory (RAM), a read-only computer memory (ROM), an erasable programmable read-only computer memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only computer memory (CD-ROM), an optical computer storage device, a magnetic computer storage device, or any suitable combination of the above.

[0507] The program product of the method for multi-target recognition in complex traffic environment of the embodiments of the present disclosure can take a portable compact disc read-only computer storage medium (CD-ROM) and include a program code, and can be run on an electronic device. However, the program product of the present disclosure is not limited thereto, and in this document, the readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus or device.

[0508] A readable signal medium can include a data signal traveling in baseband or propagated by a carrier wave appropriate for a communication network. Such a propagated signal can take a wide variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A readable signal medium can also be any readable medium that can be used to carry or store programming code for use by or in connection with an instruction execution system, apparatus, or device.

[0509] Program code embodied on a readable medium can be transmitted using any appropriate medium, including but not limited to wireless, wire line, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0510] Program code for carrying out operations of the present disclosure can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C++, etc., or conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's electronic device, partly on the user's electronic device, as a stand-alone software package, partly on the user's electronic device and partly on a remote electronic device or entirely on the remote electronic device or server. In the latter scenario, the remote electronic device can be connected to the user's electronic device through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external electronic device (for example, through the Internet using an Internet Service Provider). The application program code can be embodied in a computer readable medium, which include a data storage medium comprising physical buttons or an article of manufacture that programs an electronic device with the software.

[0511] It should be noted that although several modules of an apparatus are mentioned in the foregoing detailed description, such a division is merely exemplary and not mandatory. Indeed, features and functions of two or more modules described above can be embodied in one module, according to embodiments of the present disclosure. Conversely, features and functions of one module described above can be further divided into several modules.

[0512] Moreover, while operations of the methods of the present disclosure are described in a particular order in the drawings, this is not required or implied. That is, the operations need not be performed in the particular order described, nor do all of the illustrated operations need to be performed in order to realize the benefits of the present disclosure. Additionally or alternatively, certain steps can be omitted, combined, performed in a different order, and / or performed in parallel.

[0513] Those skilled in the art will appreciate that embodiments of the disclosure can be devised for a variety of applications. It is therefore intended that the disclosure be considered as in all respects only illustrative and not restrictive. Those skilled in the art will further appreciate that the disclosure can be used for a variety of applications. Accordingly, the disclosure is intended to embrace all alternatives, modifications and variations of the present disclosure that have been disclosed, suggested and / or can be apparent in light of the disclosure to those skilled in the art, and the present disclosure intends to embrace all alternatives, modifications and variations that fall within the scope of the claims and their equivalents. Those skilled in the art will further appreciate that the disclosure can be used for a variety of applications. Accordingly, the disclosure is intended to embrace all alternatives, modifications and variations of the present disclosure that have been disclosed, suggested and / or can be apparent in light of the disclosure to those skilled in the art, and the present disclosure intends to embrace all alternatives, modifications and variations that fall within the scope of the claims and their equivalents.

[0514] The present disclosure is described in reference to the drawings using a flowchart and / or a block diagram of the method, apparatus (system) and computer program product according to the present disclosure. It will be understood that each block of the flowchart and / or block diagram, and combinations of blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processing device or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks.

[0515] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks.

[0516] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks. Figure 1 one or more functions specified in the flowchart and / or block diagram block or blocks.

[0517] Obviously, numerous modifications and variations of the present disclosure are possible in light of the above teachings. It is therefore to be understood that within the scope of the disclosure and its equivalents, the disclosure can be practiced otherwise than as specifically described.

Claims

1. A method for multi-target recognition in complex traffic environments, characterized in that, The method includes: At specified intervals, a specified number of road point cloud data frames are acquired, wherein the road point cloud data is obtained by detecting the target road using millimeter-wave radar; For any frame of road point cloud data, the road point cloud data is projected onto the radial distance of the millimeter-wave radar to obtain the high-resolution range signal (HRRP) of the road point cloud data, and point cloud features are extracted from the road point cloud data to obtain the point cloud features of the road point cloud data; and, Based on the point cloud features, HRRP geometric features, and HRRP power spectrum features of the road point cloud data, a point cloud feature set of the road point cloud data is obtained, wherein the HRRP geometric features and the HRRP power spectrum features are obtained by feature extraction of the HRRP of the road point cloud data; Based on the point cloud feature sets of the road point cloud data for the specified number of frames, a point cloud feature sequence is obtained; The point cloud feature sequence is input into a pre-trained target classification model to obtain the category of each target object in the road point cloud data of the specified number of frames.

2. The method according to claim 1, characterized in that, The road point cloud data includes the position coordinates, radial velocity, and echo intensity of each target point; and any point cloud feature includes at least one of the following: point cloud quantity, lateral position coordinates of the point cloud centroid, longitudinal position coordinates of the point cloud centroid, ratio of lateral and longitudinal coordinates of the point cloud centroid, relative distance of the point cloud centroid to the radar, radial velocity of the point cloud centroid, variance of the radial velocity of the point cloud centroid, difference in radial velocity variation of the point cloud centroid, minimum point cloud echo intensity, maximum point cloud echo intensity, mean point cloud echo intensity, variance of point cloud echo intensity, difference in lateral position coordinates of the point cloud, difference in longitudinal position coordinates of the point cloud, variance of lateral position coordinates of the point cloud, variance of longitudinal position coordinates of the point cloud, and point cloud density. The step of extracting point cloud features from the road point cloud data to obtain the point cloud features of the road point cloud data includes: The total number of target points in the road point cloud data is determined as the point cloud quantity; and / or, The mean of the lateral position coordinates of each target point in the road point cloud data is determined as the lateral position coordinate of the centroid of the point cloud; and / or, The mean value of the ordinate of each target point in the road point cloud data is determined as the ordinate of the centroid of the point cloud; and / or, Divide the horizontal coordinate of the point cloud centroid by the vertical coordinate of the point cloud centroid to obtain the ratio of the horizontal and vertical coordinates of the point cloud centroid; and / or, Based on the horizontal and vertical coordinates of the point cloud centroid, the distance between the point cloud centroid and the radar is obtained; and / or, The average radial velocity of each target point in the road point cloud data is determined as the centroid radial velocity of the point cloud; and / or, Based on the radial velocities of each target point in the road point cloud data and the radial velocities of the point cloud centroid, the variance of the radial velocities of the point cloud centroid is obtained; and / or, The difference between the radial velocity of the target point with the largest value and the radial velocity of the target point with the smallest value in the road point cloud data is determined as the difference in radial velocity variation of the centroid of the point cloud; and / or, The echo intensity of the target point with the smallest value in the road point cloud data is determined as the minimum echo intensity value of the point cloud; and / or, The echo intensity of the target point with the largest value in the road point cloud data is determined as the maximum echo intensity of the point cloud; and / or, The average echo intensity of each target point in the road point cloud data is determined as the average echo intensity of the point cloud. The variance of the point cloud echo intensity is obtained based on the echo intensity of each target point in the road point cloud data and the mean of the point cloud echo intensity. Subtract the horizontal coordinate of the target point with the largest value from the horizontal coordinate of the smallest value in the road point cloud data to obtain the difference in the horizontal coordinate of the point cloud. The difference in the vertical coordinates of the target point with the largest value in the road point cloud data is obtained by subtracting the vertical coordinates of the target point with the smallest value. The variance of the horizontal position coordinates of the point cloud is obtained based on the horizontal position coordinates of each target point in the road point cloud data and the horizontal position coordinates of the centroid of the point cloud. The variance of the longitudinal position coordinates of the point cloud is obtained based on the longitudinal position coordinates of each target point in the road point cloud data and the longitudinal position coordinates of the centroid of the point cloud. The point cloud density is obtained based on the difference in horizontal position coordinates of the point cloud, the difference in vertical position coordinates of the point cloud, and the number of point clouds.

3. The method according to claim 2, characterized in that, The step of determining the mean of the lateral position coordinates of each target point in the road point cloud data as the lateral position coordinates of the point cloud centroid includes: The horizontal coordinates of the centroid of the point cloud can be obtained using the following formula: ； in, Let x be the x-coordinate of the centroid of the point cloud. Let i be the x-coordinate of the target point i in the road point cloud data. N is the total number of target points in the road point cloud data; The step of determining the mean of the ordinate of each target point in the road point cloud data as the ordinate of the centroid of the point cloud includes: The ordinate of the centroid of the point cloud can be obtained using the following formula: ； in, Let be the ordinate of the centroid of the point cloud. Let be the ordinate of the target point i in the road point cloud data; The step of dividing the horizontal position coordinate of the point cloud centroid by the vertical position coordinate of the point cloud centroid to obtain the ratio of the horizontal and vertical coordinates of the point cloud centroid includes: The ratio of the horizontal and vertical coordinates of the centroid of the point cloud is obtained using the following formula: ； in, The ratio of the horizontal and vertical coordinates of the centroid of the point cloud; The step of obtaining the distance between the point cloud centroid and the radar based on the horizontal and vertical coordinates of the point cloud centroid includes: The relative radar distance of the point cloud centroid is obtained using the following formula: ； in, The distance between the centroid of the point cloud and the radar. Determining the average radial velocity of each target point in the road point cloud data as the centroid radial velocity of the point cloud includes: The radial velocity of the centroid of the point cloud is obtained in the following manner: ； in, The radial velocity of the centroid of the point cloud is... Let be the radial velocity of target point i in the road point cloud data; The step of obtaining the variance of the radial velocity of the centroid of the point cloud based on the radial velocity of each target point in the road point cloud data and the radial velocity of the centroid of the point cloud includes: The variance of the radial velocity of the centroid of the point cloud is obtained using the following formula: ； in, Let Variance be the radial velocity variance of the centroid of the point cloud. The radial velocity of target point i in the road point cloud data. The radial velocity of the centroid of the point cloud; The step of determining the difference between the radial velocity of the target point with the largest value and the radial velocity of the target point with the smallest value in the road point cloud data as the difference in radial velocity change of the centroid of the point cloud includes: The difference in radial velocity variation of the centroid of the point cloud is obtained using the following formula: ； in, The difference in radial velocity variation of the centroid of the point cloud. The radial velocity of the target point with the largest value in the road point cloud data. The radial velocity of the target point with the smallest value in the road point cloud data; Determining the mean echo intensity of each target point in the road point cloud data as the mean echo intensity of the point cloud includes: The mean value of the point cloud echo intensity is obtained using the following formula: ； in, The mean value of the point cloud echo intensity. Let be the echo intensity at target point i; The step of obtaining the variance of the point cloud echo intensity based on the echo intensity of each target point in the road point cloud data and the mean of the point cloud echo intensity includes: The variance of the point cloud echo intensity is obtained using the following formula: ； in, Let Variance be the point cloud echo intensity. The echo intensity of target point i in the road point cloud data. The mean value of the point cloud echo intensity; The step of subtracting the horizontal coordinate of the target point with the largest value from the horizontal coordinate of the target point with the smallest value in the road point cloud data to obtain the difference in the horizontal coordinates of the point cloud includes: ； in, The difference in the horizontal position coordinates of the point cloud. The x-coordinate of the target point with the largest value in the road point cloud data. The x-coordinate of the target point with the smallest value in the road point cloud data; The step of subtracting the ordinate of the target point with the largest value from the ordinate of the target point with the smallest value in the road point cloud data to obtain the difference in the ordinate of the point cloud includes: The difference in the vertical position coordinates of the point cloud can be obtained using the following formula: ； in, The difference in the vertical position coordinates of the point cloud. The vertical coordinate of the target point with the largest value in the road point cloud data. The vertical coordinate of the target point with the smallest value in the road point cloud data; The step of obtaining the variance of the lateral position coordinates of the point cloud based on the lateral position coordinates of each target point in the road point cloud data and the lateral position coordinates of the point cloud centroid includes: The variance of the horizontal position coordinates of the point cloud is obtained using the following formula: ； in, Let Variance be the horizontal position coordinate variance of the point cloud. Let i be the x-coordinate of the target point i in the road point cloud data. Let x be the x-coordinate of the centroid of the point cloud; The step of obtaining the variance of the vertical position coordinates of the point cloud based on the vertical position coordinates of each target point in the road point cloud data and the vertical position coordinates of the point cloud centroid includes: The variance of the vertical position coordinates of the point cloud can be obtained using the following formula: ； in, Let Variance be the longitudinal position coordinate variance of the point cloud. Let i be the ordinate of the target point i in the road point cloud data. Let be the ordinate of the centroid of the point cloud; The step of obtaining the point cloud density based on the difference in horizontal position coordinates of the point cloud, the difference in vertical position coordinates of the point cloud, and the number of points includes: The point cloud density is obtained using the following formula: ； in, The point cloud density, The difference in the horizontal position coordinates of the point cloud. The difference in the vertical position coordinates of the point cloud is denoted as .

4. The method according to claim 1, characterized in that, The HRRP of the road point cloud data includes the amplitude of each target point in the road point cloud data; The HRRP geometric features include at least one of the following: equivalent number of scattering centers, equivalent target size, HRRP signal entropy, HRRP signal standard deviation, HRRP signal bias, HRRP signal irregularity, HRRP signal echo energy, HRRP signal energy ratio, HRRP signal skewness, and HRRP signal kurtosis. The number of equivalent scattering centers can be obtained using the following formula: ； in, The number of equivalent scattering centers. Let i be the magnitude of the target point. This is the average amplitude of each target point in the road point cloud data. The total number of target points in the road point cloud data, when hour, =1, when hour, =0; The equivalent target size is obtained using the following formula: The equivalent target size is obtained based on the position coordinates of the target point with the largest amplitude and the target point with the smallest amplitude in the HRRP. The HRRP signal entropy is obtained using the following formula: ； in, The entropy of the HRRP signal; The standard deviation of the HRRP signal is obtained using the following formula: ； in, The standard deviation of the HRRP signal; The HRRP signal deviation is obtained using the following formula: ； The HRRP signal deviation; The irregularity of the HRRP signal is obtained using the following formula: ； in, For the irregularity of the HRRP signal, Let i be the magnitude of the target i preceding the target i-1. Let i be the magnitude of the next target point i+1 after target point i; The HRRP signal echo energy is obtained using the following formula: ； in, The HRRP signal echo energy; The HRRP signal energy ratio is obtained in the following way: The amplitude values ​​of each target point in the HRRP are sorted in descending order to obtain the sorted HRRP; the sum of the squares of the amplitude values ​​of the first specified number of target points in the sorted HRRP is divided by the sum of the squares of the amplitude values ​​of all target points in the HRRP to obtain the HRRP signal energy ratio, wherein the specified number is equal to the average value of the amplitude values ​​of each target point; The HRRP signal skewness is obtained using the following formula: ； in, The HRRP signal skewness, This is the average of the absolute values ​​of the amplitudes of the target points; The HRRP signal kurtosis is obtained using the following formula: ； in, The kurtosis of the HRRP signal is given.

5. The method according to claim 1, characterized in that, The HRRP power spectrum characteristics of the road point cloud data were obtained in the following manner: For the amplitude of any target point in the HRRP of the road point cloud data, perform a Fourier transform on the amplitude of the target point to obtain the frequency domain amplitude of the target point; and, The square of the frequency domain amplitude of the target point is determined as the power spectrum of the target point; The power spectrum of each target point in the HRRP of the road point cloud data is determined as the HRRP power spectrum feature of the road point cloud data.

6. An electronic device, characterized in that, It includes a processor and a memory, which are connected via a bus; The memory stores a computer program, and the processor is configured to perform the following operations based on the computer program: At specified intervals, a specified number of road point cloud data frames are acquired, wherein the road point cloud data is obtained by detecting the target road using millimeter-wave radar; For any given frame of road point cloud data, the road point cloud data is projected onto the radial distance of the millimeter-wave radar to obtain the high-resolution range signal (HRRP) of the road point cloud data. Point cloud features are then extracted from the road point cloud data to obtain the individual point cloud features of each road point cloud data frame. Based on the point cloud features, HRRP geometric features, and HRRP power spectrum features of the road point cloud data, a point cloud feature set of the road point cloud data is obtained, wherein the HRRP geometric features and the HRRP power spectrum features are obtained by feature extraction of the HRRP of the road point cloud data; Based on the point cloud feature set of the road point cloud data of the specified number of frames, a point cloud feature sequence is obtained; The point cloud feature sequence is input into a pre-trained target classification model to obtain the category of each target object in the road point cloud data of the specified number of frames.

7. The electronic device according to claim 6, characterized in that, The road point cloud data includes the position coordinates, radial velocity, and echo intensity of each target point; and any point cloud feature includes at least one of the following: point cloud quantity, lateral position coordinates of the point cloud centroid, longitudinal position coordinates of the point cloud centroid, ratio of lateral and longitudinal coordinates of the point cloud centroid, relative distance of the point cloud centroid to the radar, radial velocity of the point cloud centroid, variance of the radial velocity of the point cloud centroid, difference in radial velocity variation of the point cloud centroid, minimum point cloud echo intensity, maximum point cloud echo intensity, mean point cloud echo intensity, variance of point cloud echo intensity, difference in lateral position coordinates of the point cloud, difference in longitudinal position coordinates of the point cloud, variance of lateral position coordinates of the point cloud, variance of longitudinal position coordinates of the point cloud, and point cloud density. The processor performs point cloud feature extraction on the road point cloud data to obtain the point cloud features of the road point cloud data, specifically configured as follows: The total number of target points in the road point cloud data is determined as the point cloud quantity; and / or, The mean of the lateral position coordinates of each target point in the road point cloud data is determined as the lateral position coordinate of the centroid of the point cloud; and / or, The mean value of the ordinate of each target point in the road point cloud data is determined as the ordinate of the centroid of the point cloud; and / or, Divide the horizontal coordinate of the point cloud centroid by the vertical coordinate of the point cloud centroid to obtain the ratio of the horizontal and vertical coordinates of the point cloud centroid; and / or, Based on the horizontal and vertical coordinates of the point cloud centroid, the distance between the point cloud centroid and the radar is obtained; and / or, The average radial velocity of each target point in the road point cloud data is determined as the centroid radial velocity of the point cloud; and / or, Based on the radial velocities of each target point in the road point cloud data and the radial velocities of the point cloud centroid, the variance of the radial velocities of the point cloud centroid is obtained; and / or, The difference between the radial velocity of the target point with the largest value and the radial velocity of the target point with the smallest value in the road point cloud data is determined as the difference in radial velocity variation of the centroid of the point cloud; and / or, The echo intensity of the target point with the smallest value in the road point cloud data is determined as the minimum echo intensity value of the point cloud; and / or, The echo intensity of the target point with the largest value in the road point cloud data is determined as the maximum echo intensity of the point cloud; and / or, The average echo intensity of each target point in the road point cloud data is determined as the average echo intensity of the point cloud. The variance of the point cloud echo intensity is obtained based on the echo intensity of each target point in the road point cloud data and the mean of the point cloud echo intensity. Subtract the horizontal coordinate of the target point with the largest value from the horizontal coordinate of the smallest value in the road point cloud data to obtain the difference in the horizontal coordinate of the point cloud. The difference in the vertical coordinates of the target point with the largest value in the road point cloud data is obtained by subtracting the vertical coordinates of the target point with the smallest value. The variance of the horizontal position coordinates of the point cloud is obtained based on the horizontal position coordinates of each target point in the road point cloud data and the horizontal position coordinates of the centroid of the point cloud. The variance of the longitudinal position coordinates of the point cloud is obtained based on the longitudinal position coordinates of each target point in the road point cloud data and the longitudinal position coordinates of the centroid of the point cloud. The point cloud density is obtained based on the difference in horizontal position coordinates of the point cloud, the difference in vertical position coordinates of the point cloud, and the number of point clouds.

8. The electronic device according to claim 7, characterized in that, The processor determines the lateral coordinates of the centroid of the point cloud by taking the average of the lateral coordinates of each target point in the road point cloud data. Specifically, it is configured as follows: The horizontal coordinates of the centroid of the point cloud can be obtained using the following formula: ； in, Let x be the x-coordinate of the centroid of the point cloud. Let i be the x-coordinate of the target point i in the road point cloud data. N is the total number of target points in the road point cloud data; The processor is specifically configured to determine the mean of the ordinates of the vertical coordinates of each target point in the road point cloud data as the ordinate of the centroid of the point cloud. The ordinate of the centroid of the point cloud can be obtained using the following formula: ； in, Let be the ordinate of the centroid of the point cloud. Let be the ordinate of the target point i in the road point cloud data; The processor executes the step of dividing the horizontal position coordinate of the point cloud centroid by the vertical position coordinate of the point cloud centroid to obtain the ratio of the horizontal and vertical coordinates of the point cloud centroid, which is specifically configured as follows: The ratio of the horizontal and vertical coordinates of the centroid of the point cloud is obtained using the following formula: ； in, The ratio of the horizontal and vertical coordinates of the centroid of the point cloud; The processor executes the step of obtaining the distance between the point cloud centroid and the radar based on the horizontal and vertical coordinates of the point cloud centroid, specifically configured as follows: The relative radar distance of the point cloud centroid is obtained using the following formula: ； in, The distance between the centroid of the point cloud and the radar. The processor is specifically configured to determine the average radial velocity of each target point in the road point cloud data as the centroid radial velocity of the point cloud. The radial velocity of the centroid of the point cloud is obtained in the following manner: ； in, The radial velocity of the centroid of the point cloud is... Let be the radial velocity of target point i in the road point cloud data; The processor executes the step of obtaining the variance of the radial velocity of the centroid of the point cloud based on the radial velocity of each target point in the road point cloud data and the radial velocity of the centroid of the point cloud, specifically configured as follows: The variance of the radial velocity of the centroid of the point cloud is obtained using the following formula: ； in, Let Variance be the radial velocity variance of the centroid of the point cloud. The radial velocity of target point i in the road point cloud data. The radial velocity of the centroid of the point cloud; The processor executes the step of determining the difference between the radial velocity of the target point with the largest value and the radial velocity of the target point with the smallest value in the road point cloud data as the difference in radial velocity change of the centroid of the point cloud, specifically configured as follows: The difference in radial velocity variation of the centroid of the point cloud is obtained using the following formula: ； Among them, among them, The difference in radial velocity variation of the centroid of the point cloud. The radial velocity of the target point with the largest value in the road point cloud data. The radial velocity of the target point with the smallest value in the road point cloud data; The processor is specifically configured to determine the mean echo intensity of each target point in the road point cloud data as the mean echo intensity of the point cloud. The mean value of the point cloud echo intensity is obtained using the following formula: ； in, The mean value of the point cloud echo intensity. Let be the echo intensity at target point i; The processor executes the step of obtaining the point cloud echo intensity variance based on the echo intensity of each target point in the road point cloud data and the mean of the point cloud echo intensity, specifically configured as follows: The variance of the point cloud echo intensity is obtained using the following formula: ； in, Let Variance be the point cloud echo intensity. The echo intensity of target point i in the road point cloud data. The mean value of the point cloud echo intensity; The processor executes the step of subtracting the horizontal coordinate of the target point with the largest value from the horizontal coordinate of the target point with the smallest value in the road point cloud data to obtain the difference in the horizontal coordinates of the point cloud, specifically configured as follows: ； in, The difference in the horizontal position coordinates of the point cloud. The x-coordinate of the target point with the largest value in the road point cloud data. The x-coordinate of the target point with the smallest value in the road point cloud data; The processor executes the step of subtracting the ordinate of the target point with the largest value from the ordinate of the target point with the smallest value in the road point cloud data to obtain the difference in the ordinate of the point cloud coordinates, specifically configured as follows: The difference in the vertical position coordinates of the point cloud can be obtained using the following formula: ； in, The difference in the vertical position coordinates of the point cloud. The vertical coordinate of the target point with the largest value in the road point cloud data. The vertical coordinate of the target point with the smallest value in the road point cloud data; The processor executes the step of obtaining the variance of the lateral position coordinates of the point cloud based on the lateral position coordinates of each target point in the road point cloud data and the lateral position coordinates of the point cloud centroid, specifically configured as follows: The variance of the horizontal position coordinates of the point cloud is obtained using the following formula: ； in, Let Variance be the horizontal position coordinate variance of the point cloud. Let i be the x-coordinate of the target point i in the road point cloud data. Let x be the x-coordinate of the centroid of the point cloud; The processor executes the step of obtaining the variance of the longitudinal position coordinates of the point cloud based on the longitudinal position coordinates of each target point in the road point cloud data and the longitudinal position coordinates of the point cloud centroid, specifically configured as follows: The variance of the vertical position coordinates of the point cloud can be obtained using the following formula: ； in, Let Variance be the longitudinal position coordinate variance of the point cloud. Let i be the ordinate of the target point i in the road point cloud data. Let be the ordinate of the centroid of the point cloud; The processor executes the step of obtaining the point cloud density based on the difference in the horizontal position coordinates of the point cloud, the difference in the vertical position coordinates of the point cloud, and the number of points in the point cloud, specifically configured as follows: The point cloud density is obtained using the following formula: ； in, The point cloud density, The difference in the horizontal position coordinates of the point cloud. The difference in the vertical position coordinates of the point cloud is denoted as .

9. The electronic device according to claim 6, characterized in that, The HRRP of the road point cloud data includes the amplitude of each target point in the road point cloud data; The HRRP geometric features include at least one of the following: equivalent number of scattering centers, equivalent target size, HRRP signal entropy, HRRP signal standard deviation, HRRP signal bias, HRRP signal irregularity, HRRP signal echo energy, HRRP signal energy ratio, HRRP signal skewness, and HRRP signal kurtosis. The processor is also configured to: The number of equivalent scattering centers can be obtained using the following formula: ； in, The number of equivalent scattering centers. Let i be the magnitude of the target point. This is the average amplitude of each target point in the road point cloud data. The total number of target points in the road point cloud data, when hour, =1, when hour, =0; The equivalent target size is obtained using the following formula: The equivalent target size is obtained based on the position coordinates of the target point with the largest amplitude and the target point with the smallest amplitude in the HRRP. The HRRP signal entropy is obtained using the following formula: ； in, The entropy of the HRRP signal; The standard deviation of the HRRP signal is obtained using the following formula: ； in, The standard deviation of the HRRP signal; The HRRP signal deviation is obtained using the following formula: ； The HRRP signal deviation; The irregularity of the HRRP signal is obtained using the following formula: ； in, For the irregularity of the HRRP signal, Let i be the magnitude of the target i preceding the target i-1. Let i be the magnitude of the next target point i+1 after target point i; The HRRP signal echo energy is obtained using the following formula: ； in, The HRRP signal echo energy; The HRRP signal energy ratio is obtained in the following way: The amplitude values ​​of each target point in the HRRP are sorted in descending order to obtain the sorted HRRP; the sum of the squares of the amplitude values ​​of the first specified number of target points in the sorted HRRP is divided by the sum of the squares of the amplitude values ​​of all target points in the HRRP to obtain the HRRP signal energy ratio, wherein the specified number is equal to the average value of the amplitude values ​​of each target point; The HRRP signal skewness is obtained using the following formula: ； in, The HRRP signal skewness, This is the average of the absolute values ​​of the amplitudes of the target points; The HRRP signal kurtosis is obtained using the following formula: ； in, The kurtosis of the HRRP signal is given.

10. The electronic device according to claim 6, characterized in that, The processor is also configured to: The HRRP power spectrum characteristics of the road point cloud data were obtained in the following manner: For the amplitude of any target point in the HRRP of the road point cloud data, perform a Fourier transform on the amplitude of the target point to obtain the frequency domain amplitude of the target point; and, The square of the frequency domain amplitude of the target point is determined as the power spectrum of the target point; The power spectrum of each target point in the HRRP of the road point cloud data is determined as the HRRP power spectrum feature of the road point cloud data.

Images