Road target classification and identification method based on millimeter wave traffic radar

Abstract

The invention discloses a road target classification and identification method based on a millimeter wave traffic radar, and the method comprises the following steps: S1, collecting the motion characteristic and scattering characteristic information of a measured target through the millimeter wave radar, obtaining the point cloud data of the measured target, and carrying out the preprocessing of the point cloud data, so as to obtain the target point cloud data; s2, performing target characteristic processing on the target point cloud data to obtain target classification data; and S3, performing posteriori and feedback on the output target classification data. The method has the advantages of being high in recognition rate and recognition accuracy and small in occupied resource and related computing power, and has good popularization prospects and application in the field of intelligent transportation.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology, and in particular to a road target classification and recognition method based on millimeter wave traffic radar. Background Art

[0002] Smart transportation is a huge intelligent system that maximizes the functions and efficiency of traffic management through perception, analysis, prediction, intervention and control. The perception system is the first link of smart transportation, helping the smart transportation system to realize functions such as traffic flow monitoring, congestion management, accident warning, etc., and improve traffic efficiency and safety. The autonomous collection of real-time traffic data information usually relies on sensor monitoring, various vehicle-mounted equipment and mobile applications. Currently, the most widely used technical means include: laser radar, high-definition camera, millimeter wave radar, microwave radar, coil, and road traffic microwave integrated detector.

[0003] Millimeter wave radar is a radar that works in the millimeter wave band. In traffic applications, it emits low-energy millimeter wave signals to the road area to be detected, and detects the real-time status by identifying the millimeter wave signals reflected by pedestrians, motor vehicles and non-motor vehicles on the road. When millimeter-wave radar is used in the intelligent infrastructure of integrated vehicle-road-cloud, the recognition and detection of vehicle trajectory information of motor vehicles, such as the target vehicle's ID, lane, length, width, speed, heading angle, etc., requires the needs of vehicle queuing detection and multi-traffic target classification and recognition computing power. The recognition efficiency is low, the recognition accuracy is low, the lane resource occupancy rate is high, and the detection computing power is complex. Summary of the invention

[0004] The present invention provides a road target classification and recognition method based on millimeter wave traffic radar with high recognition efficiency, accurate recognition degree, and low resource and computing power occupation, which can solve at least one of the above technical problems.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions: a road target classification and recognition method based on millimeter wave traffic radar is implemented by a road target classification and recognition system based on millimeter wave traffic radar, and the road target classification and recognition system based on millimeter wave traffic radar includes: a target information preprocessing module, a tracking classification module and an adaptive closed-loop module; The target information preprocessing module is used to collect and preprocess point cloud data to obtain target point cloud data. The tracking and classification module is connected to the target information preprocessing module to receive the target point cloud data and process it to obtain target classification data. The adaptive closed-loop module is connected to the tracking and classification module to perform a posteriori and feedback on the output target classification data. The following steps are also included: S1. The millimeter wave radar collects the motion characteristics and scattering characteristics information of the measured target, obtains the point cloud data of the measured target, and pre-processes the point cloud data to obtain the target point cloud data; S2, processing the target point cloud data according to target characteristics to obtain target classification data; S3. Perform a posteriori and feedback on the output target classification data.

[0006] Furthermore, the preprocessing operation in S1 further includes: S11. Eliminate data that seriously does not meet the road target speed based on traffic rules; S12, deleting noise points and false data based on the RCS characteristics of the measured target; S13, filtering data that does not conform to the vehicle movement area and movement trend based on the installation location of the millimeter wave radar and the characteristics of the static object; S14, fusing the point cloud data of 3-5 frames to increase the point cloud density.

[0007] Furthermore, the S2 further comprises: S21, obtaining target-associated point cloud information: analyzing the point cloud data associated with the target in the tracking phase, that is, traversing the tracked target, obtaining the point cloud data information associated with the current frame of the target, the point cloud data information at least including the number, RCS and motion information in spherical coordinates; S22, processing classification feature parameters, further comprising: S221, converting the acquired point cloud data information into Cartesian coordinates with the millimeter wave radar position as the origin; S222, for the target whose number of point cloud data meets the conditions, obtain the circumscribed rectangle of the point cloud data of the measured target by using the L-Shape algorithm, use the heading angle of the measured target to perform deviation correction, obtain the vertical distance between the long side of the circumscribed rectangle and the normal direction of the millimeter wave radar, and then obtain the length size of the measured target; S23, setting the classification feature structure of the candidate target: according to the data analysis results of different types of single target vehicles such as people, non-motorized vehicles, small vehicles and large vehicles, setting feature information, the feature information at least includes a classification state bit, a threshold value of the number of classifiable target point clouds, a target point cloud layer type size range and a classification execution number threshold, and the feature information is associated with the measured target at each new starting track and the feature information is initialized; S24. Construct a classifier to classify the target: traverse the current frame track target to detect each target classification status bit. 0 means that the target type has been determined and this step is skipped. Non-zero means that the target type has not been classified or the type has not been determined during the classification. For the unclassified target size, 3 frames of data are stored and the average is calculated. The average is used to compare the point cloud layer type size range. When the cumulative number of times a certain type is reached exceeds the classification execution number threshold, the type of the target is determined, and its classification status bit is set to 0. The target will no longer be classified subsequently.

[0008] Furthermore, the S3 further includes: S31, receiving the tracking targets of 3-5 frames, aggregating the targets of the same category that have been determined in the tracking process into different groups, removing the targets whose data features have too large deviations from the overall category in the category group, resetting their determined categories, and performing category discrimination again; S32. Large vehicles have long bodies, dense point clouds and are easy to split. Target splitting is likely to occur at the tracking level. If the target type is a large vehicle, a BOX restriction is set. Within the range of the BOX restriction, the tracking layer no longer starts a new track.

[0009] Compared with the prior art, the present invention has the following beneficial effects: 1. High recognition rate and accuracy: Through the sequential processing of the target information preprocessing module, the tracking and classification module and the adaptive closed-loop module, the classification of the measured target has a higher recognition rate and robustness; 2. It takes up fewer resources and involves less computing power: Compared with the combination of radar and vision classification and the classification of targets through neural networks, this technology classifies the targets during tracking on the millimeter-wave radar chip, takes up fewer resources, does not require a large amount of data training, and avoids adding additional computing power. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The drawings described herein are used to provide further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute improper limitations on the present application.

[0011] Figure 1 It is a flow chart of a method for classifying and identifying road targets based on millimeter-wave traffic radar according to an embodiment of the present invention.

[0012] Figure 2 It is a schematic block diagram of the structure of a road target classification and recognition system based on millimeter wave traffic radar according to an embodiment of the present invention.

[0013] Figure 3 is a point cloud image before preprocessing in an embodiment of the present invention.

[0014] Figure 4It is a point cloud image after preprocessing in an embodiment of the present invention.

[0015] Figure 5 It is a schematic diagram of the steps of performing target characteristic processing on target point cloud data according to an embodiment of the present invention.

[0016] Figure 6 It is a structural block diagram of a computer device according to an embodiment of the present invention.

[0017] The components in the attached figure are marked as: 1. target information preprocessing module; 2. tracking and classification module; 3. adaptive closed-loop module. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. In the absence of conflict, the embodiments in this application and the features in the embodiments can be combined with each other. Based on the embodiments in 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.

[0019] It should be noted that the meaning of "and / or" in the full text includes three parallel solutions. Taking "A and / or B" as an example, it includes solution A, solution B, or solutions that satisfy both A and B. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the protection scope required by the present invention.

[0020] See also Figure 1 The embodiment of the present invention provides a road target classification and recognition method based on millimeter wave traffic radar, comprising the following steps: S1. The millimeter wave radar collects the motion characteristics and scattering characteristics information of the measured target, obtains the point cloud data of the measured target, and pre-processes the point cloud data to obtain the target point cloud data; S2, processing the target point cloud data according to target characteristics to obtain target classification data; S3. Perform a posteriori and feedback on the output target classification data.

[0021] See also Figure 3-Figure 4 In this embodiment, the preprocessing operation in S1 further includes: S11. Eliminate data that seriously does not meet the road target speed based on traffic rules; S12, deleting noise points and false data based on the RCS characteristics of the measured target; S13, filtering data that does not conform to the vehicle movement area and movement trend based on the installation location of the millimeter wave radar and the characteristics of static objects, wherein static objects include flower beds, trees, railings, etc.; S14. Fuse 3-5 frames of the point cloud data to increase the point cloud density. Usually, the point cloud transmission frequency is 10HZ-20HZ, which does not affect the timing of the measured target data. At the same time, since the millimeter-wave radar used in traffic is usually installed on traffic lights about 8 meters away, due to the complexity of the installation location and environment, the problem of sparse point clouds is inevitable, so the number of frames needs to be fused.

[0022] Millimeter-wave radar has significant advantages in collecting information about targets in terms of the motion characteristics of objects and the scattering characteristics of radar. By emitting microwave pulses and processing the millimeter-wave signals reflected by targets on the road, the motion characteristics and scattering characteristics of the detected targets are obtained, which is usually called the point cloud information of the millimeter-wave radar. The point cloud information includes the position, speed, angle information of the detected target relative to the millimeter-wave radar, and the RCS of the target. RCS is a physical quantity that measures the intensity of the echo generated by the target under the irradiation of radar waves. The target information itself has the boundedness of motion and RCS characteristics, the correlation of the same target data or the same area data, and despite the powerful signal processing technology, there are still inevitable errors and unrecognizable complexities.

[0023] Based on this, it is necessary to pre-process the collected point cloud data to improve the data efficiency of subsequent algorithm processing of the measured target. Figure 3 As shown in, it is a point cloud image that has not been preprocessed. Figure 4 As shown, it is a pre-processed point cloud image that only retains the point cloud image related to the target vehicle under test.

[0024] See also Figure 5 In this embodiment, S2 further includes: S21, obtaining target-associated point cloud information: analyzing the point cloud data associated with the target in the tracking phase, that is, traversing the tracked target, obtaining the point cloud data information associated with the current frame of the target, the point cloud data information at least including the number, RCS and motion information in spherical coordinates, the motion information is the speed, distance, angle, etc. relative to the millimeter wave radar; S22, processing classification feature parameters, further comprising: S221, converting the acquired point cloud data information into Cartesian coordinates with the millimeter wave radar position as the origin; S222, for the target whose number of point cloud data meets the conditions, obtain the circumscribed rectangle of the point cloud data of the measured target by using the L-Shape algorithm, use the heading angle of the measured target to perform deviation correction, obtain the vertical distance between the long side of the circumscribed rectangle and the normal direction of the millimeter wave radar, and then obtain the length size of the measured target; S23, setting the classification feature structure of the candidate target: according to the data analysis results of different types of single target vehicles such as people, non-motorized vehicles, small vehicles and large vehicles, setting feature information, the feature information at least includes a classification state bit, a threshold value of the number of classifiable target point clouds, a target point cloud layer type size range and a classification execution number threshold, and the feature information is associated with the measured target at each new starting track and the feature information is initialized; S24. Construct a classifier to classify the target: traverse the current frame track target to detect each target classification status bit. 0 means that the target type has been determined and this step is skipped. Non-zero means that the target type has not been classified or the type has not been determined during the classification. For the unclassified target size, 3 frames of data are stored and the average is calculated. The average is used to compare the point cloud layer type size range. When the cumulative number of times a certain type is reached exceeds the classification execution number threshold, the type of the target is determined, and its classification status bit is set to 0. The target will no longer be classified subsequently.

[0025] In the millimeter wave radar processing stack, target classification is a post-processing step that is performed after the radar positioning processing layer. That is, after the millimeter wave radar obtains the point cloud through ADC sample processing, it obtains target tracking information and processes target characteristics through tracking, clustering, Kalman and other algorithms.

[0026] Kalman filter was once called the minimum mean square estimator. It uses noisy linear sensors (data) to minimize the mean square estimation error of linear random systems. Sensors are used to track the position of the target of interest. The target distance, speed, azimuth and other observations obtained by the sensors often contain noise. The dynamic information of the target is combined with the observation results to suppress the influence of noise and obtain a more accurate estimate of the target position, thereby achieving target tracking.

[0027] In this embodiment, S3 further includes: S31, receiving the tracking target of 3-5 frames, summarizing the targets of the same category that have been determined in the tracking process into different groups, removing the targets whose data features have too large deviation from the overall category in the category group, resetting their determined categories, and performing category discrimination again, thereby ensuring the accuracy and rigor of the classification of the measured targets; S32. Large vehicles have long bodies, dense point clouds and are easy to split. Target splitting is likely to occur at the tracking level. If the target type is a large vehicle, a BOX restriction is set. Within the range of the BOX restriction, the tracking layer no longer starts a new track to reduce target classification and improve classification accuracy.

[0028] See also Figure 2 , the embodiment of the present invention also provides a road target classification and recognition system based on millimeter wave traffic radar, including a target information preprocessing module 1, a tracking classification module 2 and an adaptive closed-loop module 3; The target information preprocessing module 1 is used to collect point cloud data and perform preprocessing to obtain target point cloud data. The tracking and classification module 2 is connected to the target information preprocessing module 1 and is used to receive the target point cloud data and process it to obtain target classification data. The adaptive closed-loop module 3 is connected to the tracking and classification module 2 and is used to perform a posteriori and feedback on the output target classification data. In this embodiment, the target information preprocessing module 1 uses a millimeter-wave radar to collect information on the motion characteristics and scattering characteristics of the target to be measured to obtain the point cloud data, and the point cloud data at least includes the position / speed / angle data of the target to be measured relative to the millimeter-wave radar, and the RCS data of the target to be measured.

[0029] In this embodiment, the target information preprocessing module 1 filters out the data in the point cloud data that does not meet the requirements of traffic regulations, vehicle motion characteristics and vehicle RCS characteristics, and performs frame fusion on the filtered and eliminated data to obtain the target point cloud data.

[0030] In this embodiment, the tracking and classification module 2 includes a Kalman prediction unit, a target association unit, a data allocation unit, a target classification unit and a data update unit; The Kalman prediction unit is used to perform a priori estimation of the trackable object based on the operation of the time n-1 state estimation; The target association unit is used to associate the point cloud data with the existing tracking target, set the wave gate function, establish the relevant centroid restriction function of the three-dimensional space, and group and associate the existing measurement points with the unique target; The data allocation unit is used to allocate the measurement values ​​to trajectories and create a measurement value group associated with each tracking target; The target classification unit is used to extract features from the associated point cloud data of each unique tracking target, including the number of associated point clouds, the distance dimension of the point cloud, the extraction and area division of the point cloud RCS features, and calculate the minimum circumscribed rectangle through the distance dimension of the associated point cloud cluster to determine the length of the target, and then determine the target category based on these features. The target classification unit needs to collect and compare data sets of different categories of measured targets, and divide the data-specific thresholds of the number of associated points, distance dimension, and target characteristic RSC of different categories; The data updating unit is used for updating the tracking target, and for each determined tracking target, updates the Kalman gain, the posterior state vector and the error operation.

[0031] An embodiment of the present invention further provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes the steps of the above-mentioned road target classification and recognition method based on millimeter-wave traffic radar.

[0032] See also Figure 6 An embodiment of the present invention further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the above-mentioned road target classification and recognition method based on millimeter-wave traffic radar.

[0033] An embodiment of the present invention further provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute the steps of the above-mentioned road target classification and recognition method based on millimeter-wave traffic radar.

[0034] It is understandable that the system, device and storage medium provided in the embodiments of the present invention correspond to the method provided in the embodiments of the present invention, and the explanation, examples and beneficial effects of the relevant contents can refer to the corresponding parts in the above methods.

[0035] It should be noted that those skilled in the art can understand that all or part of the steps implemented in the embodiments of the present invention can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using hardware, it can be implemented in whole or in part in the form of purchased standard parts or modified parts. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a DVD), or a semiconductor medium (eg, a solid state disk (SSD)).

[0036] In summary, the present application proposes a road target classification and recognition method based on millimeter-wave traffic radar. Compared with the existing technology, it has the advantages of high recognition rate and recognition accuracy, less resources occupied and less computing power involved. Through the sequential processing of the target information preprocessing module, the tracking classification module and the adaptive closed-loop module, the classification of the measured target has a higher recognition rate and robustness. At the same time, compared with the combined classification of radar and vision and the classification of the measured targets through neural networks, this technology classifies the measured targets during tracking on the millimeter-wave radar chip, occupies less resources, does not require a large amount of data training, and avoids adding additional computing power. It has good promotion prospects and application value in the field of smart transportation.

[0037] It should be understood that the examples and implementation modes described herein are for illustrative purposes only and are not intended to limit the present invention. Those skilled in the art may make various modifications or changes based on the examples and implementation modes. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A road target classification and recognition method based on millimeter wave traffic radar, characterized in that: The method is implemented by using a road target classification and recognition system based on a millimeter wave traffic radar, wherein the road target classification and recognition system based on a millimeter wave traffic radar comprises: a target information preprocessing module (1), a tracking and classification module (2) and an adaptive closed-loop module (3); The target information preprocessing module (1) is used to collect point cloud data and perform preprocessing to obtain target point cloud data; the tracking and classification module (2) is connected to the target information preprocessing module (1) and is used to receive the target point cloud data and process it to obtain target classification data; the adaptive closed-loop module (3) is connected to the tracking and classification module (2) and is used to perform a posteriori and feedback on the output target classification data; The following steps are also included: S1. The millimeter wave radar collects the motion characteristics and scattering characteristics information of the measured target, obtains the point cloud data of the measured target, and pre-processes the point cloud data to obtain the target point cloud data; S2, processing the target point cloud data according to target characteristics to obtain target classification data; S3. Perform a posteriori and feedback on the output target classification data.

2. The road target classification and recognition method based on millimeter wave traffic radar as claimed in claim 1, characterized in that: The pre-processing operation in S1 further includes: S11. Eliminate data that seriously does not meet the road target speed based on traffic rules; S12, deleting noise points and false data based on the RCS characteristics of the measured target; S13, filtering data that does not conform to the vehicle movement area and movement trend based on the installation location of the millimeter wave radar and the characteristics of the static object; S14, fusing the point cloud data of 3-5 frames to increase the point cloud density.

3. The road target classification and recognition method based on millimeter wave traffic radar as claimed in claim 1, characterized in that: The S2 further comprises: S21, obtaining target-associated point cloud information: analyzing the point cloud data associated with the target in the tracking phase, that is, traversing the tracked target, obtaining the point cloud data information associated with the current frame of the target, the point cloud data information at least including the number, RCS and motion information in spherical coordinates; S22, processing classification feature parameters, further comprising: S221, converting the acquired point cloud data information into Cartesian coordinates with the millimeter wave radar position as the origin; S222, for the target whose number of point cloud data meets the conditions, obtain the circumscribed rectangle of the point cloud data of the measured target by using the L-Shape algorithm, use the heading angle of the measured target to perform deviation correction, obtain the vertical distance between the long side of the circumscribed rectangle and the normal direction of the millimeter wave radar, and then obtain the length size of the measured target; S23, setting the classification feature structure of the candidate target: according to the data analysis results of different types of single target vehicles such as people, non-motorized vehicles, small vehicles and large vehicles, setting feature information, the feature information at least includes a classification state bit, a threshold value of the number of classifiable target point clouds, a target point cloud layer type size range and a classification execution number threshold, and the feature information is associated with the measured target at each new starting track and the feature information is initialized; S24. Construct a classifier to classify the target: traverse the current frame track target to detect each target classification status bit. 0 means that the target type has been determined and this step is skipped. Non-zero means that the target type has not been classified or the type has not been determined during the classification. For the unclassified target size, 3 frames of data are stored and the average is calculated. The average is used to compare the point cloud layer type size range. When the cumulative number of times a certain type is reached exceeds the classification execution number threshold, the type of the target is determined, and its classification status bit is set to 0. The target will no longer be classified subsequently.

4. The road target classification and recognition method based on millimeter wave traffic radar as claimed in claim 1, characterized in that: The S3 further comprises: S31, receiving the tracking targets of 3-5 frames, aggregating the targets of the same category that have been determined in the tracking process into different groups, removing the targets whose data features have too large deviations from the overall category in the category group, resetting their determined categories, and performing category discrimination again; S32. Large vehicles have long bodies, dense point clouds and are easy to split. Target splitting is likely to occur at the tracking level. If the target type is a large vehicle, a BOX restriction is set. Within the range of the BOX restriction, the tracking layer no longer starts a new track.

Images