Information processing method, device, electronic device and storage medium

Abstract

The present invention discloses an information processing method, device, electronic device and storage medium, the method comprising: obtaining information of tracks within a preset time period, obtaining information of the track to be confirmed corresponding to the preset target object, screening out the associated tracks of the track to be confirmed; dividing the associated tracks of the track to be confirmed into moving tracks and static tracks, counting the number and number of associated times of the moving tracks, the number and number of associated times of the static tracks; constructing an associated feature matrix of the track to be confirmed based on the number and number of associated times of the moving tracks, the number and number of associated times of the static tracks, and the existence time of the track to be confirmed; based on the associated feature matrix, when the preset target object is a false target, deleting all information of the track to be confirmed. The present invention improves the accuracy of false target identification and elimination from the dimensions of high scene coverage of false target identification on road conditions, rich mathematical and statistical basis, and low algorithm level, and effectively reduces the possibility of false alarms and missed alarms generated by radar.

Description

Technical Field

[0001] The present invention relates to the field of millimeter wave radar detection, and in particular to an information processing method, device, electronic equipment and storage medium. Background Art

[0002] Millimeter-wave radar has a certain probability of forming false targets during vehicle driving, causing the radar to produce false alarms and missed alarms, affecting the performance of the radar system. As for the elimination of false targets, related technologies can only suppress false targets in a single case, and the scene coverage rate is low. Summary of the invention

[0003] In order to solve the problems of the prior art, the embodiments of the present invention provide an information processing method, device, electronic device and storage medium. The technical solution is as follows:

[0004] In one aspect, an information processing method is provided, the method comprising:

[0005] Obtaining the information of the point trace within a preset time period; the point trace refers to the energy information generated by the echo of the radar electromagnetic wave irradiating the preset target object; the point trace information includes speed and position;

[0006] Based on the information of the point track, the information of the track to be confirmed corresponding to the preset target object is obtained;

[0007] Based on the information of the track to be confirmed and the information of the point track, the associated point track of the track to be confirmed is screened out;

[0008] The associated points of the track to be confirmed are divided into moving points and static points, and the number and the number of associated points of the moving points, and the number and the number of associated points of the static points are counted;

[0009] Based on the number and correlation times of moving tracks, the number and correlation times of static tracks, and the existence time of the tracks to be confirmed, a correlation feature matrix of the tracks to be confirmed is constructed;

[0010] Based on the correlation feature matrix, determine whether the preset target is a false target; if so, delete all information of the track to be confirmed; if not, retain all information of the track to be confirmed.

[0011] In another aspect, an information processing device is provided, the device comprising:

[0012] The point trace acquisition module is used to obtain the information of the point trace within a preset time period; the point trace refers to the energy information generated by the echo of the radar electromagnetic wave irradiating the preset target object; the point trace information includes speed and position;

[0013] A to-be-confirmed track acquisition module is used to obtain the information of the to-be-confirmed track corresponding to the preset target object based on the information of the point track;

[0014] The associated point track screening module is used to screen out the associated point tracks of the track to be confirmed based on the information of the track to be confirmed and the information of the point tracks;

[0015] The associated point track classification and statistics module is used to classify the associated point tracks of the track to be confirmed into moving point tracks and static point tracks, and to count the number and number of associated points of the moving point tracks, and the number and number of associated points of the static point tracks;

[0016] A correlation feature matrix construction module is used to construct a correlation feature matrix of the track to be confirmed based on the number and correlation times of the moving track, the number and correlation times of the static track, and the existence time of the track to be confirmed;

[0017] The false target judgment module is used to judge whether the preset target is a false target based on the associated feature matrix; if so, all information of the track to be confirmed is deleted; if not, all information of the track to be confirmed is retained.

[0018] In an exemplary embodiment, the to-be-confirmed track acquisition module includes:

[0019] The initial track acquisition module is used to cluster points with similar speed and position to obtain the initial track;

[0020] The initial track processing module is used to process each initial track through the Kalman filter algorithm and output the initial track that meets the preset conditions as the track to be confirmed.

[0021] In an exemplary embodiment, the associated point trace screening module includes:

[0022] A matrix construction module, used for constructing a track state matrix and a point track information matrix based on the information of the track to be confirmed and the information of the point track;

[0023] The correlation statistics acquisition module is used to obtain the correlation statistics between the track to be confirmed and the point track based on the track state matrix and the point track information matrix;

[0024] The associated point track determination module is used to determine that any point track is an associated point track of the track to be confirmed when the associated statistic between any point track and the track to be confirmed is less than a preset associated threshold.

[0025] In an exemplary embodiment, the associated point trace classification statistics module includes:

[0026] The number statistics module is used to count the number of moving traces and the number of static traces;

[0027] The association times statistics module is used to increase the association times of the motion traces by 1 when the number of the motion traces is greater than the preset motion threshold; and to increase the association times of the static traces by 1 when the number of the static traces is greater than the preset static threshold.

[0028] In an exemplary embodiment, the false target determination module includes:

[0029] A false target probability acquisition module is used to obtain the false target probability corresponding to the track to be confirmed based on the correlation feature matrix;

[0030] The false target determination module is used to determine that the preset target object is a false target when the false target probability is greater than a preset false target probability threshold.

[0031] In an exemplary embodiment, the false target probability acquisition module includes:

[0032] The native false track probability acquisition module is used to obtain the native false track probability based on the correlation feature matrix; the native false track probability refers to the false target probability corresponding to the uncorrected track to be confirmed;

[0033] The native false track probability correction module is used to correct the native false track probability through smoothing calculation to obtain the false target probability corresponding to the track to be confirmed.

[0034] In an exemplary embodiment, the native false track probability acquisition module includes:

[0035] The correlation feature matrix display module is used to display the correlation feature matrix;

[0036] A labeling information acquisition module, used to obtain labeling information corresponding to the associated feature matrix in response to the labeling operation on the associated feature matrix; the labeling information indicates information of the real target object during the actual road test;

[0037] The training module is used to input the associated feature matrix and the corresponding annotation information into the machine learning model for training to obtain the support vector machine model parameter set and global bias;

[0038] The native false track probability output module is used to obtain the native false track probability based on the correlation feature matrix, the support vector machine model parameter set and the global bias.

[0039] On the other hand, an electronic device is provided, including a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the above-mentioned information processing method.

[0040] On the other hand, a computer-readable storage medium is provided, in which at least one instruction or at least one program is stored, and the at least one instruction or the at least one program is loaded and executed by a processor to implement the information processing method as described above.

[0041] On the other hand, a computer program product or a computer program is provided, the computer program product or the computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the electronic device performs the information processing method provided in the above aspects.

[0042] The embodiment of the present invention obtains the information of the to-be-confirmed track corresponding to the preset target object by acquiring the information of the track within a preset time period, and screens out the associated track of the to-be-confirmed track; divides the associated track of the to-be-confirmed track into moving track and static track, and counts the number of moving track and the number of associated times, and the number of static track and the number of associated times; constructs the associated feature matrix of the to-be-confirmed track based on the number of moving track and the number of associated times, the number of static track and the number of associated times, and the existence time of the to-be-confirmed track; based on the associated feature matrix, when the preset target object is a false target, deletes all the information of the to-be-confirmed track, and improves the accuracy of false target recognition and elimination from the dimensions of high scene coverage of false target recognition under road conditions, rich mathematical and statistical basis, and low algorithm level, and effectively reduces the possibility of false alarms and missed alarms by radar.

[0043] The attached figure shows

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0045] Figure 1 It is a flowchart of an information processing method provided by an embodiment of the present invention;

[0046] Figure 2 is a flowchart of another information processing method provided by an embodiment of the present invention;

[0047] Figure 3 is a structural block diagram of an information processing device provided by an embodiment of the present invention;

[0048] Figure 4 This is a hardware structure block diagram of a terminal provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0050] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices.

[0051] See also Figure 1 , which shows a flow chart of an information processing method provided by an embodiment of the present invention. It should be noted that this specification provides method operation steps as described in the embodiment or flow chart, but more or fewer operation steps may be included based on conventional or non-creative labor. The order of steps listed in the embodiment is only one way of executing the steps among many orders, and does not represent the only order of execution. When the system or product is executed in practice, it can be executed in the order of the method shown in the embodiment or the accompanying drawings or in parallel (for example, in a parallel processor or multi-threaded processing environment). Specifically, Figure 1 As shown, the method may include:

[0052] S101, obtaining information of point traces within a preset time period.

[0053] Among them, the dot trace refers to the energy information generated by the echo of the radar electromagnetic wave irradiating the preset target object. The dot trace information can include speed and position, as well as azimuth, signal-to-noise ratio (SIGNAL-NOISE RATIO, SNR), radar cross section (RADAR CROSS SECTION, RCS), etc. Specifically, the position can be the distance of the dot trace relative to the radar of the vehicle. The preset target objects can be four-wheeled vehicles, two-wheeled vehicles, pedestrians, metal guardrails, etc. near the vehicle. Except for metal guardrails, which are completely static objects, vehicles and pedestrians have dynamic and static states. Optionally, the preset time period is an algorithm cycle, such as 50 milliseconds.

[0054] S103, based on the information of the point track, obtaining the information of the to-be-confirmed track corresponding to the preset target object.

[0055] The information of the track to be confirmed may include speed and position.

[0056] S105, based on the information of the track to be confirmed and the information of the point track, filter out the associated point tracks of the track to be confirmed.

[0057] S107, dividing the associated points of the track to be confirmed into moving points and static points, and counting the number of moving points and the number of associated times, and the number of static points and the number of associated times.

[0058] The moving traces may be traces with a speed greater than or equal to 1 m / s, and the static traces may be traces with a speed less than 1 m / s. D The number of associations of the moving points is recorded as T D The number of static points is denoted as N S The number of correlations of static points is denoted as T S .

[0059] S109, constructing a correlation feature matrix of the unconfirmed track based on the number and correlation times of the moving track, the number and correlation times of the static track, and the existence time of the unconfirmed track.

[0060] As can be seen from the "number of associations plus 1" in S107, this method will go through a cycle of "one algorithm cycle after another" when processing information. The existence time of the track to be confirmed is related to the number of algorithm cycles experienced, specifically, how many algorithm cycles the track to be confirmed has been retained, recorded as Age. Obviously, in the first algorithm cycle, Age is 1.

[0061] The correlation feature matrix is ​​recorded as M, and the calculation formula of M is:

[0062] M=[m1,m2,m3,m4] (1)

[0063] in,

[0064] S111, based on the correlation feature matrix, determine whether the preset target is a false target; if so, execute step S113; otherwise, if not, execute step S115.

[0065] S113, deleting all information of the track to be confirmed.

[0066] S115, retain all information of the track to be confirmed.

[0067] The embodiment of the present invention identifies and removes false targets after the point trace acquisition is completed. Compared with the false target identification and removal in the point trace acquisition stage, the probability of missing the real target is reduced to a certain extent, and the accuracy of false target identification is improved. In addition, the embodiment of the present invention has no restrictions on the state of the preset target, and can handle false targets generated under various working conditions, which significantly improves the scene coverage.

[0068] In an exemplary embodiment, the above step S103 obtains information of the to-be-confirmed track corresponding to the preset target object based on the information of the point track, and may include the following steps:

[0069] Cluster the points with similar speed and position to obtain the initial track;

[0070] Each initial track is processed by the Kalman filter algorithm, and the initial track that meets the preset conditions is output as the track to be confirmed. Specifically, it will go through three stages internally.

[0071] The first stage is point clustering.

[0072] Cluster the traces obtained in S101 to generate multiple trace cluster sample sets. The so-called "clustering" means taking a trace as a root point and finding traces with similar speed and position around the root point. The number of traces in the generated trace sample set is not fixed. There may be 5 traces in this trace sample set and 6 traces in that trace sample set. Optionally, the clustering algorithm uses DBSCAN (Density-Based Spatial Clustering of Applications with Noise).

[0073] The second stage is track initialization.

[0074] Not all initial tracks generated at this stage will be output as tracks to be confirmed. How to determine which initial tracks can be output as tracks to be confirmed can be confirmed using mn logic. In the first algorithm cycle of mn logic, the point track sample set is fitted to generate the initial track through the least squares method. The initial track can be a rectangular frame. The information of the initial track can include speed and position, as well as heading angle, track frame length, track frame width, etc. In terms of quantitative relationship, the initial track may be generated by fitting a single point track sample in a single point track sample set, or it may be generated by fitting some / all point tracks in multiple point track sample sets. After the initial track is generated, the ID of the initial track (such as track 1, track 2, track 3) is obtained, and the track initialization is completed.

[0075] The third stage is to wait for confirmation of track judgment.

[0076] For each initial track, the judgment method is the same. Take track 1 as an example: obtain the information of the track of the second algorithm cycle; through the Kalman filter algorithm, traverse the correlation between the track of the second algorithm cycle and track 1, find the track of the second algorithm cycle that is associated with track 1, that is, obtain the associated track of track 1; use the associated track to maintain and update track 1, that is, based on the speed, position and other information of the associated track, update the information of track 1, such as position and speed; after the update is completed, continue to use the Kalman filter algorithm to update and maintain the information of the initial track of the previous algorithm cycle with the information of the track of the next algorithm cycle. When the internal setting of the mn logic is "for a single initial track, 6 algorithm cycles out of 10 algorithm cycles can be associated with the track, then the initial track can be output as a track to be confirmed", if track 1 meets this condition, track 1 can be output as a track to be confirmed, and the mn logic ends. That is, the internal setting of the mn logic is the preset condition for outputting the track to be confirmed.

[0077] In the embodiment of the present invention, the output of the track to be confirmed includes three stages: point track clustering, track initialization, and track to be confirmed judgment. Compared with directly outputting the fitting result after point track clustering, the "best of the best" setting of the present invention makes the track to be confirmed more consistent with the actual state of the preset target, thereby improving the accuracy of false target recognition, and further reducing the possibility of false alarms and missed alarms by the radar.

[0078] In an exemplary embodiment, the above step S105 filters out the associated track points of the track to be confirmed based on the information of the track to be confirmed and the information of the track points, and may include the following steps:

[0079] Based on the information of the track to be confirmed and the information of the point track, a track state matrix and a point track information matrix are constructed;

[0080] Based on the track state matrix and the point track information matrix, the correlation statistics between the track to be confirmed and the point track are obtained;

[0081] The information of all track points is traversed, and when the correlation statistic between any track point and the track to be confirmed is less than a preset correlation threshold, any track point is determined to be a correlation track of the track to be confirmed.

[0082] As mentioned above, the information of the track to be confirmed may include speed and position. The track state matrix of the track to be confirmed at time k is recorded as The calculation formula is:

[0083]

[0084] Wherein, k indicates which algorithm cycle is currently in, for example, k=1 indicates that it is currently in the first algorithm cycle, and k=2 indicates that it is currently in the second algorithm cycle; X posis the longitudinal distance of the track; X v Y is the longitudinal velocity of the track; pos Y is the lateral distance of the track; v is the lateral velocity of the track.

[0085] The information of the trace can include speed and position. The trace information matrix at time k is recorded as (k). Taking a trace m as an example, the trace information matrix Z at time k is m The calculation formula of (k) is:

[0086] Z m (k) = [X m|pos , X m|v , Y m|pos , Y m|v ] (3)

[0087] Among them, X m|pos is the longitudinal distance of point m; X m|v Y is the longitudinal velocity of point m; m|pos Y is the horizontal distance of point m; m|v is the lateral velocity of point m.

[0088] After obtaining the track state matrix And the point information matrix Z m (k), the correlation statistic between the track to be confirmed and the point track m is recorded as V k , V k The calculation formula is:

[0089]

[0090] Among them, S(k) is the innovation covariance in the Kalman filtering process.

[0091] If the correlation statistic V between the track to be confirmed and the point track m k If the value is less than the preset correlation threshold, it is determined that the point m is the associated point of the track to be confirmed. The preset correlation threshold is obtained by querying the chi-square distribution table. Optionally, the preset correlation threshold is 3. For each point, the processing method is the same. Therefore, after the point traversal is completed, all the associated points of the track to be confirmed are screened out.

[0092] The embodiment of the present invention provides a rigorous mathematical and statistical basis for determining the "associated point traces", and the associated point traces are the basis for subsequent judgment of false targets, that is, the rigorous mathematical and statistical basis in the early stage improves the accuracy of the final identification and elimination of false targets.

[0093] In an exemplary embodiment, the counting of the number of moving traces and the number of associations, and the number of static traces and the number of associations in the above step S107 may include the following steps:

[0094] Count the number of moving traces and the number of static traces;

[0095] When the number of moving traces is greater than the preset motion threshold, the number of associations of the moving traces is increased by 1; when the number of static traces is greater than the preset static threshold, the number of associations of the static traces is increased by 1.

[0096] Specifically, "the number of associations plus 1" reflects that the information processing of this method is a cyclic process of "one algorithm cycle after another". For track 1, the number of associations of the moving point track is T D , the number of associations of static points T S The initial value of is 0. In the first algorithm cycle, if the number of moving points N D The number of static points N that are greater than the preset motion threshold S If it is greater than the preset static threshold, the number of associations of the moving traces is T D , the number of associations of static points T S All become 1. As can be seen from S111, the last step of this method is to decide whether to delete the track information. Therefore, if track 1 is retained in the first algorithm cycle, it will enter the second algorithm cycle. In the second algorithm cycle, if the number of moving point tracks N D The number of static points N that are greater than the preset motion threshold S If it is greater than the preset static threshold, the number of associations of the moving traces is T D , the number of associations of static points T S As long as track 1 exists in the algorithm cycle, the number of associations of the moving track T D , the number of associations of static points T S It will continue to accumulate.

[0097] The preset motion threshold and the preset static threshold can be set according to the distance between the track to be confirmed and the ego vehicle. For example, for a track within 20 meters from the ego vehicle, the preset motion threshold can be 5 and the preset static threshold can be 4. That is, for a track, in the current algorithm cycle, if the associated motion point tracks are greater than 5, then the number of associated motion point tracks T D If the number of static traces associated is greater than 4, then the number of static traces associated is T. S Just add 1.

[0098] The embodiment of the present invention divides the associated traces into motion traces and static traces, and the preset motion threshold and the preset static threshold are set according to the actual situation, so as to be closer to the real scene. If the motion traces and static traces are not distinguished, and the preset motion threshold and the preset static threshold are not distinguished, and they are directly processed together, the information processing efficiency will be reduced, which is not conducive to the final identification and elimination of false targets.

[0099] In an exemplary embodiment, judging whether the preset target object is a false target based on the correlation feature matrix in the above step S111 may include:

[0100] Based on the correlation feature matrix, the false target probability corresponding to the track to be confirmed is obtained.

[0101] When the false target probability is greater than a preset false target probability threshold, the preset target object is determined to be a false target. Optionally, the preset false target probability threshold is 0.8.

[0102] The embodiment of the present invention calculates probability through a matrix and then judges false targets based on the probability, with simple steps, clear logic and low algorithm complexity.

[0103] In an exemplary embodiment, obtaining the false target probability corresponding to the to-be-confirmed track based on the correlation feature matrix may include:

[0104] Based on the correlation feature matrix, the original false track probability is obtained, where the original false track probability refers to the false target probability corresponding to the uncorrected track to be confirmed.

[0105] Through smoothing calculation, the original false track probability is corrected to obtain the false target probability corresponding to the track to be confirmed. Generally speaking, smoothing calculation includes smoothing filtering technology.

[0106] The original false track probability is denoted as The false target probability is denoted as P F|k , P F|k The calculation formula is:

[0107]

[0108] Where α and β are probability smoothing coefficients, which are empirical values. Generally, α is defined as 0.8 and β is defined as 0.2. F|k-1 is the false target probability of the previous algorithm cycle, that is, the meaning of formula (5) is: combined with the false target probability P at the previous moment F|k-1 and the native false target probability at the current moment Calculate the false target probability P at the current moment F|k Obviously, in the first algorithm cycle, the false target probability P of the previous algorithm cycle is F|k-1is 0. If the track to be confirmed is retained in S111 of the first algorithm cycle, then after entering the second cycle, the false target probability P of the previous algorithm cycle is F|k-1 is the false target probability P of the first algorithm cycle F|k .

[0109] The embodiment of the present invention uses smoothing calculation to correct the probability value. Compared with directly using the probability result based on the output of the correlation feature matrix, using the corrected probability result is obviously more accurate.

[0110] In an exemplary embodiment, obtaining the native false track probability based on the correlation feature matrix may include:

[0111] Display the correlation feature matrix.

[0112] Specifically, it can be displayed in a visualization tool script, which can display the track information and aircraft frame in real-time video to facilitate subsequent manual labeling operations.

[0113] In response to the labeling operation on the correlation feature matrix, labeling information corresponding to the correlation feature matrix is ​​obtained.

[0114] The annotation information indicates the information of the real target object in the actual road test process. When manually annotating based on the road test information, if there is no target object in the real scene of the road test, but a pending track appears in the visualization tool script, the pending track will be marked as a false target track. It should be noted that annotation is only required in the first algorithm cycle. After the subsequent machine learning model is learned, no annotation is required.

[0115] The associated feature matrix and the corresponding annotation information are input into the machine learning model for training to obtain the support vector machine model parameter set and global bias.

[0116] For machine learning models, the associated feature matrix and the corresponding annotation information are training samples; and the support vector machine model parameter set includes the feature mean set, feature standard deviation set and feature weight set, which are denoted as Mμ, Sigma and Beta respectively, and are specifically expressed as:

[0117] Mμ=[Mμ1, Mμ2,..., Mμ n ] (6)

[0118] Sigma=[Sigma1, Sigma2,...,Sigma n ] (7)

[0119] Beta=[Beta1, Beta2,...,Beta n] (8)

[0120] Wherein, n is the feature dimension of the training sample. In the embodiment of the present invention, n is related to the number of elements of the associated feature matrix, that is, n is 4.

[0121] Based on the correlation feature matrix, support vector machine model parameter set and global bias, the native false track probability is obtained.

[0122] The global bias is recorded as Bias, the original false track probability The calculation formula is:

[0123]

[0124] Among them, Score is a process parameter, which is jointly determined by the associated feature matrix M, the feature mean set Mμ, the feature standard deviation set Sigma, the feature weight set Beta and the global bias Bias. The specific calculation formula is:

[0125]

[0126] Based on the process parameter Score, in addition to being able to calculate the probability of native false tracks In addition, the original true target probability can be calculated, that is, the true target probability corresponding to the uncorrected track to be confirmed. The original true target probability is recorded as The calculation formula is:

[0127]

[0128] It can be seen from the support vector machine model parameter set that the machine learning algorithm used in the embodiment of the present invention is the support vector machine (SVM) algorithm. In fact, machine learning includes many algorithms, which are somewhat substitutable for each other. However, the technical problems solved by different algorithms and the storage resources and computing power resources required are not the same. For example, the decision tree algorithm can also determine whether the preset target is a false target, but it cannot provide a rigorous mathematical basis for the calculation of the probability of false targets; although the perceptron algorithm can output the probability of false targets, the storage resources required by the algorithm are greater than the support vector machine algorithm used in the embodiment of the present invention. Therefore, the information processing method based on the support vector machine provided in the embodiment of the present invention has the characteristics of clear structure, high accuracy, low algorithm complexity, and can be easily and quickly transplanted to the existing algorithm as an independent operation unit.

[0129] See also Figure 2 , which is a flow chart of another information processing method provided by an embodiment of the present invention, the method comprising:

[0130] S201, obtaining information of point traces within a preset time period.

[0131] S203, clustering points with similar speed and position to obtain an initial track.

[0132] S205, processing each initial track by using a Kalman filter algorithm, and outputting the initial track that meets the preset conditions as the track to be confirmed.

[0133] S207, constructing a track state matrix and a point track information matrix based on the information of the track to be confirmed and the information of the point track.

[0134] S209, based on the track state matrix and the point track information matrix, obtain the correlation statistics between the track to be confirmed and the point track.

[0135] S211, traversing the information of all track points, when the correlation statistic between any track point and the track to be confirmed is less than a preset correlation threshold, determining that any track point is a correlation track of the track to be confirmed.

[0136] S213, dividing the associated points of the track to be confirmed into moving points and static points, and counting the number of moving points and the number of static points.

[0137] S215, when the number of motion traces is greater than the preset motion threshold, the number of associations of the motion traces is increased by 1; when the number of static traces is greater than the preset static threshold, the number of associations of the static traces is increased by 1.

[0138] S217, displays the correlation feature matrix.

[0139] S219, in response to the labeling operation on the associated feature matrix, obtaining labeling information corresponding to the associated feature matrix.

[0140] S221, input the associated feature matrix and the corresponding annotation information into the machine learning model for training to obtain the support vector machine model parameter set and global bias.

[0141] S223, based on the correlation feature matrix, the support vector machine model parameter set and the global bias, the native false track probability is obtained.

[0142] S225, correcting the original false track probability through smoothing calculation to obtain the false target probability corresponding to the track to be confirmed.

[0143] S227, when the false target probability is greater than the preset false target probability threshold, determine whether the preset target is a false target; if so, execute step S229; otherwise, if not, execute step S531.

[0144] S229, deleting all information of the track to be confirmed.

[0145] S231, retain all information of the track to be confirmed.

[0146] The specific implementation of steps S201 to S231 can refer to the relevant steps in the aforementioned method embodiment, which will not be repeated here.

[0147] Corresponding to the information processing methods provided in the above-mentioned embodiments, an embodiment of the present invention further provides an information processing device. Since the information processing device provided in the embodiment of the present invention corresponds to the information processing methods provided in the above-mentioned embodiments, the implementation methods of the aforementioned information processing methods are also applicable to the information processing device provided in this embodiment and will not be described in detail in this embodiment.

[0148] See also Figure 3 , which is a schematic diagram of the structure of an information processing device provided by an embodiment of the present invention. The information processing device 300 has the function of implementing the information processing method in the above method embodiment. The function can be implemented by hardware or by hardware executing corresponding software. Figure 3 As shown, the information processing device 300 may include:

[0149] The point trace acquisition module 310 is used to obtain the information of the point trace within a preset time period; the point trace refers to the energy information generated by the echo of the radar electromagnetic wave irradiating the preset target object; the point trace information includes speed and position;

[0150] The to-be-confirmed track acquisition module 320 is used to obtain the information of the to-be-confirmed track corresponding to the preset target object based on the information of the point track;

[0151] The associated point track screening module 330 is used to screen out the associated point tracks of the track to be confirmed based on the information of the track to be confirmed and the information of the point tracks;

[0152] The associated point track classification and statistics module 340 is used to classify the associated point tracks of the track to be confirmed into moving point tracks and static point tracks, and to count the number and number of associated points of the moving point tracks, and the number and number of associated points of the static point tracks;

[0153] The correlation feature matrix construction module 350 is used to construct the correlation feature matrix of the track to be confirmed based on the number and correlation times of the moving track, the number and correlation times of the static track, and the existence time of the track to be confirmed;

[0154] The false target judgment module 360 ​​is used to judge whether the preset target is a false target based on the correlation feature matrix; if so, all information of the track to be confirmed is deleted; if not, all information of the track to be confirmed is retained.

[0155] In an exemplary embodiment, the to-be-confirmed track acquisition module 320 may include:

[0156] The initial track acquisition module is used to cluster points with similar speed and position to obtain the initial track;

[0157] The initial track processing module is used to process each initial track through the Kalman filter algorithm and output the initial track that meets the preset conditions as the track to be confirmed.

[0158] In an exemplary embodiment, the associated point trace screening module 330 may include:

[0159] A matrix construction module, used for constructing a track state matrix and a point track information matrix based on the information of the track to be confirmed and the information of the point track;

[0160] The correlation statistics acquisition module is used to obtain the correlation statistics between the track to be confirmed and the point track based on the track state matrix and the point track information matrix;

[0161] The associated point track determination module is used to traverse the information of all point tracks, and when the associated statistic between any point track and the track to be confirmed is less than a preset associated threshold, determine that any point track is an associated point track of the track to be confirmed.

[0162] In an exemplary embodiment, the associated point trace classification statistics module 340 may include:

[0163] The number statistics module is used to count the number of moving traces and the number of static traces;

[0164] The association times statistics module is used to increase the association times of the motion traces by 1 when the number of the motion traces is greater than the preset motion threshold; and to increase the association times of the static traces by 1 when the number of the static traces is greater than the preset static threshold.

[0165] In an exemplary embodiment, the false target determination module 360 ​​may include:

[0166] A false target probability acquisition module is used to obtain the false target probability corresponding to the track to be confirmed based on the correlation feature matrix;

[0167] The false target determination module is used to determine that the preset target object is a false target when the false target probability is greater than a preset false target probability threshold.

[0168] In an exemplary embodiment, the false target probability acquisition module may include:

[0169] The native false track probability acquisition module is used to obtain the native false track probability based on the correlation feature matrix; the native false track probability refers to the false target probability corresponding to the uncorrected track to be confirmed;

[0170] The native false track probability correction module is used to correct the native false track probability through smoothing calculation to obtain the false target probability corresponding to the track to be confirmed.

[0171] In an exemplary embodiment, the native false track probability acquisition module may include:

[0172] The correlation feature matrix display module is used to display the correlation feature matrix;

[0173] A labeling information acquisition module, used to obtain labeling information corresponding to the associated feature matrix in response to the labeling operation on the associated feature matrix; the labeling information indicates information of the real target object during the actual road test;

[0174] The training module is used to input the associated feature matrix and the corresponding annotation information into the machine learning model for training to obtain the support vector machine model parameter set and global bias;

[0175] The native false track probability output module is used to obtain the native false track probability based on the correlation feature matrix, the support vector machine model parameter set and the global bias.

[0176] It should be noted that the device provided in the above embodiment, when implementing its functions, only uses the division of the above functional modules as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0177] The information processing device of the embodiment of the present invention achieves high scene coverage of false target identification in road conditions, rich mathematical and statistical basis, low algorithm level and other dimensions, thereby improving the accuracy of false target identification and elimination and effectively reducing the possibility of false alarms and missed alarms generated by radar.

[0178] An embodiment of the present invention provides a terminal, which includes a processor and a memory, in which at least one instruction, at least one program, a code set or an instruction set is stored, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the information processing method provided in the above method embodiment.

[0179] The memory can be used to store software programs and modules. The processor executes various functional applications and information processing by running the software programs and modules stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, application programs required for functions, etc.; the data storage area can store data created according to the use of the device, etc. In addition, the memory can include a high-speed random access memory and can also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory can also include a memory controller to provide the processor with access to the memory.

[0180] The method embodiments provided in the embodiments of the present invention can be executed in a computer terminal, a server or a similar computing device. Taking running on a terminal as an example, Figure 4 : is a hardware structure block diagram of a terminal running an information processing method provided by an embodiment of the present invention, specifically:

[0181] The terminal may include components such as an RF (Radio Frequency) circuit 410, a memory 420 including one or more computer-readable storage media, an input unit 430, a display unit 440, a sensor 450, an audio circuit 460, a WiFi (wireless fidelity) module 470, a processor 480 including one or more processing cores, and a power supply 490. Those skilled in the art will appreciate that Figure 4 The terminal structure shown in the figure does not constitute a limitation on the terminal, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0182] The RF circuit 410 can be used for receiving and sending signals during information transmission or calls. In particular, after receiving the downlink information of the base station, it is handed over to one or more processors 480 for processing; in addition, the data related to the uplink is sent to the base station. Generally, the RF circuit 410 includes but is not limited to an antenna, at least one amplifier, a tuner, one or more oscillators, a user identity module (SIM) card, a transceiver, a coupler, an LNA (Low Noise Amplifier), a duplexer, etc. In addition, the RF circuit 410 can also communicate with the network and other terminals through wireless communication. The wireless communication can use any communication standard or protocol, including but not limited to GSM (Global System of Mobile communication), GPRS (General Packet Radio Service), CDMA (Code Division Multiple Access), WCDMA (Wideband Code Division Multiple Access), LTE (Long Term Evolution), email, SMS (Short Messaging Service), etc.

[0183] The memory 420 can be used to store software programs and modules, and the processor 480 executes various functional applications and data processing by running the software programs and modules stored in the memory 420. The memory 420 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, application programs required for functions, etc.; the data storage area may store data created according to the use of the terminal, etc. In addition, the memory 420 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory 420 may also include a memory controller to provide the processor 480 and the input unit 430 with access to the memory 420.

[0184] The input unit 430 can be used to receive input digital or character information, and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control. Specifically, the input unit 430 may include a touch-sensitive surface 431 and other input devices 432. The touch-sensitive surface 431, also known as a touch display screen or a touchpad, can collect user touch operations on or near it (such as operations performed by users using fingers, styluses, or any other suitable objects or accessories on or near the touch-sensitive surface 431), and drive corresponding connection devices according to a pre-set program. Optionally, the touch-sensitive surface 431 may include a touch detection device and a touch controller. Among them, the touch detection device detects the user's touch orientation, detects the signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts it into touch point coordinates, and then sends it to the processor 480, and can receive and execute commands sent by the processor 480. In addition, the touch-sensitive surface 431 can be implemented in various types such as resistive, capacitive, infrared, and surface acoustic waves. In addition to the touch-sensitive surface 431, the input unit 430 may further include other input devices 432. Specifically, other input devices 432 may include but are not limited to one or more of a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, a joystick, and the like.

[0185] The display unit 440 can be used to display information input by the user or information provided to the user and various graphical user interfaces of the terminal, which can be composed of graphics, text, icons, videos and any combination thereof. The display unit 440 may include a display panel 441, and optionally, the display panel 441 may be configured in the form of LCD (Liquid Crystal Display), OLED (Organic Light-Emitting Diode), etc. Further, the touch-sensitive surface 431 may cover the display panel 441, and when the touch-sensitive surface 431 detects a touch operation on or near it, it is transmitted to the processor 480 to determine the type of the touch event, and then the processor 480 provides corresponding visual output on the display panel 441 according to the type of the touch event. Among them, the touch-sensitive surface 431 and the display panel 441 can be two independent components to realize the input and output functions, but in some embodiments, the touch-sensitive surface 431 and the display panel 441 can also be integrated to realize the input and output functions.

[0186] The terminal may also include at least one sensor 450, such as a light sensor, a motion sensor, and other sensors. Specifically, the light sensor may include an ambient light sensor and a proximity sensor, wherein the ambient light sensor may adjust the brightness of the display panel 441 according to the brightness of the ambient light, and the proximity sensor may turn off the display panel 441 and / or the backlight when the terminal is moved to the ear. As a type of motion sensor, the gravity acceleration sensor can detect the magnitude of acceleration in each direction (generally three axes), and can detect the magnitude and direction of gravity when stationary. It can be used for applications that identify the terminal posture (such as horizontal and vertical screen switching, related games, magnetometer posture calibration), vibration recognition related functions (such as pedometer, tapping), etc.; as for other sensors such as gyroscopes, barometers, hygrometers, thermometers, infrared sensors, etc. that can be configured in the terminal, they will not be repeated here.

[0187] The audio circuit 460, the speaker 461, and the microphone 462 can provide an audio interface between the user and the terminal. The audio circuit 460 can transmit the received audio data to the speaker 461 after converting the received audio data into an electrical signal, which is converted into a sound signal for output by the speaker 461; on the other hand, the microphone 462 converts the collected sound signal into an electrical signal, which is received by the audio circuit 460 and converted into audio data, and then the audio data is output to the processor 480 for processing, and then sent to another terminal through the RF circuit 410, or the audio data is output to the memory 420 for further processing. The audio circuit 460 may also include an earplug jack to provide communication between an external headset and the terminal.

[0188] WiFi is a short-range wireless transmission technology. The terminal can help users send and receive emails, browse web pages, and access streaming media through the WiFi module 470, which provides users with wireless broadband Internet access. Figure 4 A WiFi module 470 is shown, but it is understandable that it is not an essential component of the terminal and can be omitted as needed without changing the essence of the invention.

[0189] The processor 480 is the control center of the terminal, and uses various interfaces and lines to connect various parts of the entire terminal. By running or executing software programs and / or modules stored in the memory 420, and calling data stored in the memory 420, the processor 480 performs various functions of the terminal and processes data, thereby monitoring the terminal as a whole. Optionally, the processor 480 may include one or more processing cores; preferably, the processor 480 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application programs, and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 480.

[0190] The terminal also includes a power supply 490 (such as a battery) for supplying power to each component. Preferably, the power supply can be logically connected to the processor 480 through a power management system, so that the power management system can manage charging, discharging, and power consumption management. The power supply 490 can also include any components such as one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, and power status indicators.

[0191] Although not shown, the terminal may also include a camera, a Bluetooth module, etc., which will not be described in detail herein. Specifically in this embodiment, the terminal also includes a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by one or more processors. The one or more programs include instructions for executing the information processing provided by the above method embodiment.

[0192] An embodiment of the present invention also provides a computer-readable storage medium, which can be set in a terminal to store at least one instruction, at least one program, a code set or an instruction set related to implementing an information processing method. The at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the information processing method provided by the above-mentioned method embodiment.

[0193] Optionally, in this embodiment, the above-mentioned storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store program codes.

[0194] It should be noted that the order of the above embodiments of the present invention is for description only and does not represent the superiority or inferiority of the embodiments. The above describes a specific embodiment of the present invention. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0195] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0196] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0197] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An information processing method, characterized in that: include: Get the information of points within a preset time period; The point trace refers to the energy information generated by the echo of the radar electromagnetic wave irradiating the preset target object; The information of the point trace includes speed and position; Based on the information of the point track, obtaining the information of the to-be-confirmed track corresponding to the preset target object; Based on the information of the track to be confirmed and the information of the point track, selecting the associated point track of the track to be confirmed; Dividing the associated points of the to-be-confirmed track into moving points and static points, and counting the number and the number of associated points of the moving points, and the number and the number of associated points of the static points; Based on the number and association times of the moving tracks, the number and association times of the static tracks, and the existence time of the to-be-confirmed tracks, construct an association feature matrix of the to-be-confirmed tracks; Based on the correlation feature matrix, determining whether the preset target object is a false target; If yes, all the information of the unconfirmed track is deleted; if no, all the information of the unconfirmed track is retained.

2. The information processing method according to claim 1, characterized in that: The obtaining, based on the information of the point track, the information of the to-be-confirmed track corresponding to the preset target object includes: Clustering the points with similar speed and position to obtain an initial track; Each of the initial tracks is processed by a Kalman filter algorithm, and an initial track that meets a preset condition is output as the track to be confirmed.

3. The information processing method according to claim 1, characterized in that: The step of screening out associated point tracks of the to-be-confirmed track based on the information of the to-be-confirmed track and the information of the point tracks comprises: Based on the information of the track to be confirmed and the information of the point track, construct a track state matrix and a point track information matrix; Based on the track state matrix and the point track information matrix, obtaining the correlation statistics between the track to be confirmed and the point track; When the correlation statistic between any of the track points and the track to be confirmed is less than a preset correlation threshold, it is determined that any of the track points is a correlation point track of the track to be confirmed.

4. The information processing method according to claim 1, characterized in that: The counting of the number and the number of associations of the moving traces, and the number and the number of associations of the static traces, includes: Counting the number of the moving traces and the number of the static traces; When the number of the moving traces is greater than the preset motion threshold, the number of associations of the moving traces is increased by 1; when the number of the static traces is greater than the preset static threshold, the number of associations of the static traces is increased by 1.

5. The information processing method according to claim 1, characterized in that: The determining whether the preset target object is a false target based on the association feature matrix includes: Based on the correlation feature matrix, obtaining a false target probability corresponding to the to-be-confirmed track; When the false target probability is greater than a preset false target probability threshold, the preset target object is determined to be a false target.

6. The information processing method according to claim 5, characterized in that: The obtaining, based on the correlation feature matrix, a false target probability corresponding to the to-be-confirmed track includes: Based on the correlation feature matrix, a native false track probability is obtained; the native false track probability refers to the false target probability corresponding to the uncorrected track to be confirmed; The original false track probability is corrected through smoothing calculation to obtain the false target probability corresponding to the track to be confirmed.

7. The information processing method according to claim 6, characterized in that: The method of obtaining the probability of a native false track based on the correlation feature matrix includes: displaying the correlation feature matrix; In response to the labeling operation on the correlation feature matrix, labeling information corresponding to the correlation feature matrix is ​​obtained; the labeling information indicates information of a real target object in an actual drive test process; Inputting the associated feature matrix and the corresponding annotation information into a machine learning model for training to obtain a support vector machine model parameter set and a global bias; The native false track probability is obtained based on the associated feature matrix, the support vector machine model parameter set and the global bias.

8. An information processing device, characterized in that: The device comprises: A point trace acquisition module is used to acquire point trace information within a preset time period; the point trace refers to the energy information generated by the echo of the radar electromagnetic wave irradiating the preset target object; the point trace information includes speed and position; A to-be-confirmed track acquisition module, used to obtain information of the to-be-confirmed track corresponding to the preset target object based on the information of the point track; A related point track screening module, used for screening out the related point tracks of the track to be confirmed based on the information of the track to be confirmed and the information of the point track; The associated point track classification and statistics module is used to divide the associated point tracks of the to-be-confirmed track into moving point tracks and static point tracks, and to count the number and number of associated points of the moving point tracks, and the number and number of associated points of the static point tracks; A correlation feature matrix construction module, used to construct a correlation feature matrix of the track to be confirmed based on the number and correlation times of the moving track, the number and correlation times of the static track, and the existence time of the track to be confirmed; The false target judgment module is used to judge whether the preset target is a false target based on the associated feature matrix; if so, delete all information of the track to be confirmed; if not, retain all information of the track to be confirmed.

9. An electronic device, characterized in that: It comprises a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the information processing method according to any one of claims 1 to 7.

10. A computer-readable storage medium, wherein at least one instruction or at least one program is stored in the computer-readable storage medium, and the at least one instruction or the at least one program is loaded and executed by a processor to implement the information processing method according to any one of claims 1 to 7.

Images