Lane identification method and device, electronic equipment and medium

Abstract

The embodiment of the invention discloses a lane recognition method and device, electronic equipment and a medium. The method comprises the following steps: if a target is detected through a radar antenna, determining detection data of the radar antenna for the target; determining a lane identifier of a lane where the target is located, and taking the detection data and the lane identifier as sample data; and performing model training according to the sample data to determine a lane recognition model, and performing lane recognition based on the lane recognition model. According to the scheme, lane recognition is directly carried out according to the most direct detection data of the radar antenna for the target instead of carrying out lane recognition after the specific position of the target is calculated according to the detection data of the radar antenna, so that the problem that lane recognition is carried out according to the specific position of the target can be solved; and the problem of low lane recognition accuracy caused by target position detection errors is solved, so that the lane recognition accuracy is improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular, to a lane recognition method, apparatus, electronic device, and medium. Background Art

[0002] As an all-weather and all-day perception device, millimeter-wave radar has inherent advantages of high ranging and speed measurement accuracy and is widely used. However, due to the high cost of millimeter-wave radar antennas and receivers, it is impossible to use a large number of antennas to measure the target angle, resulting in low target angle measurement accuracy and seriously affecting the application range of millimeter-wave radar. The large-aperture antenna array technology is applied to traffic millimeter-wave radar in order to greatly improve the angle measurement accuracy on the premise of maintaining the cost.

[0003] However, in the application of large-aperture antenna arrays, DOA (Direction Of Arrival) estimation methods all assume that the target is in the far field of the radar antenna. When the target is in the far field of the radar antenna, the signal phase received by the radar antenna is a linear function of the position of the radar antenna, and DOA estimation can be based on the plane wave model. However, for large-aperture antennas, the far-field conditions are more stringent. In the case of using a large-aperture antenna for target detection, the target is equivalent to being in the near field of the radar antenna. At this time, the signal phase received by the radar antenna is not a linear function of the position of the radar antenna. When using the curved wave model for DOA estimation, the difficulty and error of model establishment and solution are relatively large, thereby affecting the accuracy of lane recognition. Summary of the Invention

[0004] Embodiments of this application provide a lane recognition method, apparatus, electronic device, and medium, so as to perform lane recognition based on the detection data of the radar antenna for the target and improve the accuracy of lane recognition.

[0005] According to one aspect of this application, a lane recognition method is provided. The method includes:

[0006] If a target is detected by a radar antenna, determine the detection data of the radar antenna for the target;

[0007] Determine the lane identifier of the lane where the target is located, and use the detection data and the lane identifier as sample data;

[0008] Perform model training based on the sample data to determine a lane recognition model, and perform lane recognition based on the lane recognition model.

[0009] According to one aspect of this application, a lane recognition apparatus is provided. The apparatus includes:

[0010] A detection data determination module, configured to determine the detection data of the radar antenna for a target if the target is detected by the radar antenna;

[0011] A sample data determination module, configured to determine the lane identifier of the lane where the target is located, and use the detection data and the lane identifier as sample data;

[0012] A model training module, configured to perform model training based on the sample data to determine a lane recognition model, so as to perform lane recognition based on the lane recognition model.

[0013] According to another aspect of the present application, an electronic device is provided, and the electronic device includes:

[0014] At least one processor; and

[0015] A memory communicatively connected to the at least one processor; wherein,

[0016] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the lane recognition method of any embodiment of the present application.

[0017] According to another aspect of the present application, a computer-readable storage medium is provided, and the computer-readable storage medium stores computer instructions for causing a processor to implement the lane recognition method of any embodiment of the present application when executed.

[0018] In the technical solution of the embodiment of the present application, if a target is detected by a radar antenna, the detection data of the radar antenna for the target is determined; the lane identifier of the lane where the target is located is determined, and the detection data and the lane identifier are used as sample data; model training is performed based on the sample data to determine a lane recognition model, so as to perform lane recognition based on the lane recognition model. The above solution can perform lane recognition directly based on the most direct detection data of the radar antenna for the target, rather than calculating the specific position of the target based on the detection data of the radar antenna and then performing lane recognition, which can solve the problem of low lane recognition accuracy caused by the detection error of the target position in the solution of lane recognition based on the specific position of the target, thereby improving the accuracy of lane recognition.

[0019] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present application, nor is it used to limit the scope of the present application. Other features of the present application will become easily understood through the following description. Description of the Drawings

[0020] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on these drawings.

[0021] Figure 1 is a flowchart of a lane recognition method provided in Embodiment 1 of the present application;

[0022] Figure 2 is a flowchart of a lane recognition method provided in Embodiment 2 of the present application;

[0023] Figure 3 is a first schematic diagram of a lane intercepting line provided in Embodiment 2 of the present application;

[0024] Figure 4 is a second schematic diagram of a lane intercepting line provided in Embodiment 2 of the present application;

[0025] Figure 5 is a schematic diagram of a 3D radar provided in Embodiment 2 of the present application;

[0026] Figure 6 is a schematic structural diagram of a lane recognition device provided in Embodiment 4 of the present application;

[0027] Figure 7 is a schematic structural diagram of an electronic device provided in Embodiment 5 of the present application. Detailed implementation manners

[0028] To enable those skilled in the art to better understand the solutions of the present application, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0029] It should be noted that the terms "first", "second", "third", "fourth", "actual", "preset", etc. in the description, claims and the above-mentioned drawings of this application are used to distinguish similar objects, and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of this application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0030] Embodiment 1

[0031] Figure 1 FIG. is a flowchart of a lane recognition method provided for Embodiment 1 of this application. The embodiments of this application are applicable to the situation of lane recognition. Typically, it is applicable to the situation of lane recognition based on the detection data of a target by a radar antenna. This method can be executed by a lane recognition device, which can be implemented in the form of hardware and / or software, and the lane recognition device can be configured in an electronic device. As Figure 1 shown, this method includes:

[0032] S110. If a target is detected by the radar antenna, determine the detection data of the radar antenna for the target.

[0033] Among them, the radar can be of any type, such as a centimeter-wave radar, a millimeter-wave radar, a microwave radar, etc., and can be set in any scenario. Typically, the embodiments of this application are applicable to the case where the radar antenna is a large-aperture radar antenna, so as to solve the problem of low accuracy of the conventional lane recognition scheme due to the characteristics of the large-aperture radar antenna. The large-aperture radar antenna can be, for example, a radar antenna with an aperture of more than 100 millimeters. The type of the target is not limited and can be any dynamic or static target, such as a vehicle, a person, a stone, etc. The type of the target to be detected can be preset according to the actual situation. The detection data of the radar antenna for the target can include the echo data of the detection wave emitted by the radar antenna reflected by the target, and can also include other data calculated based on the echo data.

[0034] Exemplarily, if there is no target within the detection range of the radar antenna, the detection wave emitted by the radar antenna will not be reflected, and the radar antenna will not receive echo data. If a target is detected through the radar antenna, it means that the radar antenna has received echo data, and the detection data of the radar antenna for this target can be determined. During the process of the radar determining the detection data for this target, it is necessary to track the target to clarify the detection data for the same target and group the detection data for the same target. Specifically, when a target is detected, the echo signal received by the radar antenna can be processed to determine the target distance between the target and the radar antenna, and based on the change in the target distance and the detection frequency of the radar antenna, the speed of the target can be determined. The target is tracked according to the speed of the target, and the targets with unchanged speed are determined as the same target, or the targets with continuously changing speed are determined as the same target.

[0035] S120. Determine the lane identifier of the lane where the target is located, and use the detection data and the lane identifier as sample data.

[0036] Among them, the lane identifier can be a lane number, a lane position, a lane line number, etc. The lane identifier of the lane where the target is located can be recognized by an image collector, or manually input, or reported by the target itself.

[0037] Exemplarily, the detection data can be used as feature data, and the lane identifier can be used as a label to form sample data for model training. The sample data can be divided according to a ratio. One part is used as training data for model training, and one part is used as test data for testing the trained model and evaluating the performance of the model. For example, 70% of the sample data can be used as training samples, and 30% of the sample data can be used as test samples.

[0038] S130. Perform model training based on the sample data to determine a lane recognition model, and perform lane recognition based on the lane recognition model.

[0039] Exemplarily, the lane recognition model can be determined based on model training using sample data, so as to facilitate subsequent lane recognition based on the lane recognition model and the detection data collected by the radar antenna. Specifically, during the model training process, the sample data used for model training can be input into a trainer for training. The trainer can be a trainer based on algorithms such as Naive Bayes, Logistic Regression, Decision Tree, Random Forest, Support Vector Machine, etc. The kernel function can be selected as a linear function to simplify the functional relationship. Grid search can be used for optimization iteration to search for the optimal parameters and construct the lane recognition model. After the lane recognition model training is completed, the lane recognition model can be tested using the sample data for testing, and the accuracy, precision, recall rate, F1 score, etc. of the lane recognition model can be evaluated. If the indicators of the lane recognition model do not meet the requirements, additional sample data can be added for continuous training and optimization until the indicators meet the requirements.

[0040] In the application scenario using a large-aperture radar antenna, the target is equivalent to being in the near field of the radar antenna. The phase of the signal received by the radar antenna is not linearly related to the position of the radar antenna. The deflection angle of the target relative to the radar antenna depends on the relationship between the signal phase and the position of the radar antenna. There are errors in the modeling and solution of the relationship between the signal phase and the position of the radar antenna, resulting in errors in the deflection angle of the target relative to the radar antenna. Consequently, there are errors in the lateral distance of the target relative to the radar antenna calculated based on the deflection angle, and the accuracy of lane recognition based on the lateral distance is low. Here, the lateral distance is the distance perpendicular to the lane direction. However, the solution of the embodiment of the present application does not perform lane recognition based on the lateral distance, so there is no need to model and solve the relationship between the signal phase and the position of the radar antenna, and no errors are introduced in lane recognition. There is a corresponding relationship between the lane identifier and the lateral distance, and the lateral distance can be obtained by solving the channel data, and the channel data can reflect the lateral distance. In the recognition of the lane identifiers of lanes where targets with different target distances are located, the target distance and the channel data reflecting the lateral distance can be used as training data to learn the relationship between the training data and the lane identifiers through model learning. Using the detection data and the lane identifiers as sample data for model training, the relationship between the detection data and the lane number can be learned more accurately through the model, improving the accuracy of lane recognition.

[0041] In the technical solution of the embodiment of the present application, if a target is detected by a radar antenna, the detection data of the radar antenna for the target is determined; the lane identifier of the lane where the target is located is determined, and the detection data and the lane identifier are used as sample data; a lane recognition model is determined through model training based on the sample data, so as to perform lane recognition based on the lane recognition model. The above solution can perform lane recognition directly according to the most direct detection data of the radar antenna for the target, rather than calculating the specific position of the target based on the detection data of the radar antenna and then performing lane recognition, which can solve the problem of low lane recognition accuracy caused by the detection error of the target position in the solution of performing lane recognition according to the specific position of the target, thereby improving the accuracy of lane recognition.

[0042] Embodiment 2

[0043] Figure 2 The following is a flowchart of a lane recognition method provided by the second embodiment of the present application. The second embodiment of the present application is optimized based on the above embodiment. For the solutions not described in detail in the second embodiment of the present application, please refer to the above embodiment. As Figure 2 shown, the method of the second embodiment of the present application specifically includes the following steps:

[0044] S210. If a target is detected by a radar antenna, the target distance between the target and the radar antenna is detected by the radar antenna.

[0045] S220. If the target distance reaches a preset distance, the detection data of the radar antenna for the target is determined.

[0046] Exemplarily, sometimes it may be only necessary to perform lane recognition at a preset position point, and the distance and direction deviation angle between the preset position point and the radar antenna are known. Or only recognize the lane where the target passing through a certain preset line is located. As Figure 3 shown, the lane where the target passing through the intercept line can be recognized. This situation is generally applied to the scenario of traffic flow detection. When detecting the traffic flow passing through the intercept line on each lane, it is necessary to recognize the lane number of the lane where the vehicle passing through the intercept line is located, so as to distinguish the traffic flow on different lanes. The preset distance can be determined according to the preset position point, and the preset distance is the distance between the preset position point and the radar antenna. Or the preset distance is determined according to the intercept line, indicating the distance between the intercept line and the radar antenna. If the target distance between the target and the radar antenna reaches the preset distance, the detection data of the radar antenna for the target is determined to determine the detection data of the radar antenna for the target when the target reaches the preset position point or the intercept line.

[0047] In the application of the embodiments of the present application, generally, the distance between the intercept line and the preset position points on the same intercept line relative to the radar antenna is more than 50 meters. In the case of a relatively long distance, it can be considered that the distance of an intercept line relative to the radar antenna is consistent, and the distances of the preset position points on the same intercept line relative to the radar antenna are consistent, corresponding to a preset distance.

[0048] In the embodiments of the present application, if the target distance reaches the preset distance, the detection data of the radar antenna for the target is determined, including:

[0049] If it is detected in the target frame that the target distance reaches the preset distance, the detection data of the radar antenna for the target within a preset time traced forward from the target frame is determined.

[0050] Exemplarily, in order to enrich the sample data for model training, the determined detection data of the radar antenna for the target may not be just the detection data of one frame, but the detection data corresponding to multiple frames. The preset time can be determined in advance. If it is detected in the target frame that the target distance reaches the preset distance, that is, the target moves to the preset position point or intercept line, the detection data of the radar antenna for the target within the preset time traced forward from the target frame is determined. It is defaulted that the target moves in a straight line within the preset time and does not change lanes. Therefore, the labels corresponding to the detection data of the radar antenna for the target determined within the preset time period are the same, all being the same lane identifier. Specifically, the preset time can be the duration corresponding to 5 frames of detection data, that is, 5 consecutive frames of detection data are determined each time. The preset time can also be greater than the duration corresponding to 5 frames of detection data. 5 consecutive or non-consecutive frames of detection data can be selected from the preset time, and the data other than the 5 frames within the preset time is used as impurity data and not stored. During the actual storage of the detection data, the detection data can be stored in a structure in chronological order. The structure can be a structure capable of storing 5 frames of detection data. The structure can be a "first-in, first-out" structure, that is, the first stored detection data can be taken out first, and this structure can meet the requirement of updating according to time and retain the latest detection data. For example Figure 3As shown, the target moves closer to the radar antenna along the moving direction, and the target distance from the radar antenna becomes smaller and smaller. An identifier FLAG = 0 can be set to indicate that the target distance has not reached the preset distance, that is, the target has not reached the interception line. In this state, during the process of the radar antenna detecting and tracking the target, each time a frame of detection data for the target is determined, it is stored in the structure. When the number of detection data in the structure reaches 5 frames, and when the detection data is collected and stored in the structure again, the earliest stored detection data in the structure is pushed out, so as to retain the latest 5 frames of detection data. When it is detected in the target frame that the target distance reaches the preset distance, the identifier FLAG = 1 is set, the detection data of the radar antenna for the target determined in the target frame is stored in the structure, the earliest stored detection data in the structure is pushed out, and the process of continuing to store the detection data is stopped. The detection data stored in the structure at this time is used as the complete detection data of the radar antenna for the target.

[0051] In the embodiment of the present application, if the preset distance is one, the detection data includes the channel data of the radar antenna; if the preset distance is at least two, the detection data includes the target distance and the channel data. Among them, the channel data includes the echo data that the detection wave emitted by the radar antenna is reflected by the target and then returned and received by the radar antenna, including the characteristics such as the amplitude, frequency, and phase of the echo data. The lane identifier of the lane where the target is located mainly corresponds to the lateral distance of the target relative to the radar. The lateral distance can be calculated based on the target distance and the DOA. The DOA can be calculated through the channel data. That is, there is a certain relationship between the lane identifier, the target distance, and the channel data. The relationship between the lane identifier, the target distance, and the channel data can be learned through the model to perform lane recognition.

[0052] Exemplarily, if the trained model is used to identify the lane where the target passing through a position point or passing through an interception line is located, that is, the preset distance is one, the target distance of the target can be regarded as a fixed constant. At this time, only the relationship between the lane identifier and the channel data when the radar antenna detects the target needs to be learned through the model, and the target distance does not need to be added. If the preset distance is at least two, as Figure 4 shown, there are at least two interception lines, and the preset distances between different interception lines and the radar antenna are different. Then the model needs to learn the relationship between the lane identifier and the two detection data of the channel data and the target distance when the radar antenna detects the target. Therefore, the detection data needs to include the target distance and the channel data.

[0053] In the embodiment of the present application, the determination process of the channel data includes:

[0054] If a target is detected by the radar antenna, determine the target position of the target;

[0055] Determine the channel data returned from the target position collected by the radar antenna.

[0056] Exemplarily, as Figure 5 shown, if the radar antenna detects a target, a three-dimensional radar map is established, where the range dimension represents the distance between the target and the radar antenna, the Doppler dimension represents the velocity information of the target, and the antenna dimension represents the channel data actually received by the radar antenna. The target position of the target can be determined according to the data in the range dimension and the Doppler dimension, so as to determine the channel data returned from the target position collected by the radar antenna, which represents the channel data returned by the detected target, thereby more accurately determining the detection data for the target.

[0057] S230. Determine the lane identifier of the lane where the target is located, and use the detection data and the lane identifier as sample data.

[0058] S240. Perform model training according to the sample data to determine a lane recognition model, and perform lane recognition based on the lane recognition model.

[0059] In the embodiment of the present application, the method further includes:

[0060] If a preset object is detected by the radar antenna, determine the detection data of the radar antenna for the preset object;

[0061] Input the detection data of the radar antenna for the preset object into the lane recognition model to determine the lane identifier of the lane where the preset object is located.

[0062] Among them, the preset object can be determined according to the actual situation, such as a vehicle, a person, a stone, etc. Exemplarily, in the application process of the lane recognition model, a preset object is detected by the radar antenna. If the preset object is detected, the detection data of the radar antenna for the preset object is determined. The type of the detection data of the radar antenna for the preset object is the same as that of the detection data of the radar antenna for the target. If the detection data of the radar antenna for the target includes the channel data of the radar antenna and the target distance of the target relative to the radar antenna, the detection data of the radar antenna for the preset object also includes the channel data of the radar antenna and the distance of the preset object relative to the radar antenna. If the detection data of the radar antenna for the target only includes the channel data of the radar antenna, the detection data of the radar antenna for the preset object also only includes the channel data of the radar antenna. Input the detection data of the radar antenna for the preset object into the lane recognition model, and the lane identifier of the lane where the preset object is located is output through the lane recognition model to achieve lane recognition.

[0063] In an embodiment of the present application, if a preset object is detected by a radar antenna, the detection data of the radar antenna for the preset object is determined, including:

[0064] If the distance between the preset object detected by the radar antenna and the radar antenna reaches a preset distance, the detection data of the radar antenna for the preset object is determined.

[0065] Exemplarily, sometimes it may only be necessary to perform lane recognition on a preset position point, or only on the lane where a preset object passing through a line is located. For example, Figure 3 As shown, lane recognition can be performed on the lane where a preset object passing through the intercepted line is located. The preset distance can be determined according to the preset position point or the intercepted line, and the preset distance represents the distance between the position point or the intercepted line and the radar antenna. If the distance between the preset object detected by the radar antenna and the radar antenna reaches the preset distance, the detection data of the radar antenna for the preset object is determined to identify the lane number of the lane where the preset object passing through the preset position point or the intercepted line is located.

[0066] An embodiment of the present application provides a lane recognition method. The target distance between the target and the radar antenna is detected by the radar antenna; if the target distance reaches the preset distance, the detection data of the radar antenna for the target is determined. The lane identifier of the lane where the target is located is determined, and the detection data and the lane identifier are used as sample data; model training is performed according to the sample data to determine a lane recognition model, so as to perform lane recognition based on the lane recognition model. The above solution can specifically identify the lane where the target with a target distance reaching the preset distance is located, and does not need to rely on the specific position of the target for recognition, avoiding the problem of low lane recognition accuracy caused by position calculation errors, and improving the accuracy of lane recognition.

[0067] Embodiment III

[0068] An embodiment of the present application is a specific implementation manner of the above embodiment. The embodiment of the present application is optimized based on the above embodiment, and the solutions not described in detail in the embodiment of the present application can be seen in the above embodiment. The method of the embodiment of the present application specifically includes the following steps:

[0069] S310. Extract the target distance r in the target track information and the channel data when the radar antenna detects the target and save them in structure A. Among them, structure A stores the sample data for inputting into the classifier, and can store K frames of data at the same time. Typically, K = 5 can be taken. The channel data collected by the radar antenna is complex, and the real part and the imaginary part are stored separately when stored as features.

[0070] S3: The structure A is a first-in-first-out structure. When the target gradually approaches the intercept line, the distance r from the intercept line becomes smaller and smaller. Set an identifier FLAG = 0, indicating that the target has not passed the intercept line. When FLAG = 0, the target distance r and channel data of the latest frame replace the data of the earliest frame in A, and keep the information in A arranged according to the frame number.

[0071] S4: When the target reaches the intercept line, update the target distance r and channel data x in A for the last time, and set FLAG = 1.

[0072] S5: Identify the lane identifier of the lane where the target is located at the intercept line through manual observation or automatic camera recognition, and save the lane identifier in A. The information in A at this time is the sample data finally input to the trainer.

[0073] S6: A total of M samples are collected. Randomly select a part of the samples (for example, 70%), and input the data in A into the SVM trainer for training. The kernel function can be selected as a linear kernel, and the grid search method is used for optimization. After being processed by the SVM trainer, obtain the functional relationship between the lane identifier and the target distance r and channel data : If the preset distance is 1, then obtain the functional relationship between the lane identifier and the channel data after being processed by the SVM trainer :

[0074] S7: Use the remaining samples (for example, 30%) as the test set to test the function f. When the test index meets the business requirements, the lane recognition algorithm f can be determined. Otherwise, repeat S1 - S6 to retrain by increasing the sample size until the business requirements are met.

[0075] Embodiment 4

[0076] Figure 6 FIG. is a schematic structural diagram of a lane recognition device provided in Embodiment 4 of the present application. The device can execute the lane recognition method provided in any embodiment of the present application, and has corresponding functional modules and beneficial effects for executing the method. As Figure 6 shown, the device includes:

[0077] A detection data determination module 410, configured to determine the detection data of the radar antenna for the target if a target is detected through the radar antenna;

[0078] A sample data determination module 420, configured to determine the lane identifier of the lane where the target is located, and use the detection data and the lane identifier as sample data;

[0079] The model training module 430 is configured to perform model training based on the sample data to determine a lane recognition model, so as to perform lane recognition based on the lane recognition model.

[0080] In an embodiment of the present application, the detection data determination module 410 includes:

[0081] The target distance determination unit is configured to detect the target distance between the target and the radar antenna through the radar antenna;

[0082] The data determination unit is configured to determine the detection data of the radar antenna for the target if the target distance reaches a preset distance.

[0083] In an embodiment of the present application, the data determination unit includes:

[0084] The judgment subunit is configured to determine the detection data of the radar antenna for the target within a preset time period traced forward from the target frame if the target distance is detected to reach the preset distance in the target frame.

[0085] In an embodiment of the present application, if the preset distance is one, the detection data includes the channel data of the radar antenna; if the preset distance is at least two, the detection data includes the target distance and the channel data.

[0086] In an embodiment of the present application, the device further includes:

[0087] The target position determination module is configured to determine the target position of the target if a target is detected through the radar antenna;

[0088] The channel data determination module is configured to determine the channel data collected by the radar antenna and returned from the target position.

[0089] In an embodiment of the present application, the device further includes:

[0090] The preset object recognition module is configured to determine the detection data of the radar antenna for the preset object if a preset object is detected through the radar antenna;

[0091] The lane recognition module is configured to input the detection data of the radar antenna for the preset object into the lane recognition model to determine the lane identifier of the lane where the preset object is located.

[0092] In an embodiment of the present application, the preset object recognition module includes:

[0093] The preset object detection unit is configured to determine the detection data of the radar antenna for the preset object if the distance between the preset object and the radar antenna is detected to reach the preset distance through the radar antenna.

[0094] A lane recognition device provided by an embodiment of the present application can execute a lane recognition method provided by any embodiment of the present application, and has functional modules and beneficial effects corresponding to the execution of the method.

[0095] Embodiment Five

[0096] Figure 7 FIG. shows a schematic structural diagram of an electronic device 10 that can be used to implement an embodiment of the present application. The electronic device is intended to represent various forms of digital computers, such as, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, personal digital processors, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present application described herein and / or claimed.

[0097] As Figure 7 shown, the electronic device 10 includes at least one processor 11, and a memory communicatively connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc. Among them, the memory stores a computer program executable by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. The input / output (I / O) interface 15 is also connected to the bus 14.

[0098] Multiple components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disc, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0099] The processor 11 may be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the lane recognition method.

[0100] In some embodiments, the lane recognition method may be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed onto the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the lane recognition method described above may be executed. Alternatively, in other embodiments, the processor 11 may be configured to execute the lane recognition method by any other suitable means (e.g., by means of firmware).

[0101] The various embodiments of the systems and techniques described above herein may be implemented in digital electronic circuitry, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include: implemented in one or more computer programs that may be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a special-purpose or general-purpose programmable processor that receives data and instructions from a storage system, at least one input device, and at least one output device, and transmits the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0102] The computer programs for implementing the methods of the present application may be written in any combination of one or more programming languages. These computer programs may be provided to the processors of a general-purpose computer, a special-purpose computer, or other programmable lane recognition devices such that when the computer programs are executed by the processors, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer programs may be executed entirely on the machine, partially on the machine, as a stand-alone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0103] In the context of this application, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium can be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0104] To provide for interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can also be used to provide for interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0105] The systems and techniques described herein can be implemented in a computing system that includes backend components (such as, for example, a data server), or a computing system that includes middleware components (such as, for example, an application server), or a computing system that includes frontend components (such as, for example, a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (such as, for example, a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0106] A computing system may include a client and a server. The client and the server are generally far from each other and usually interact via a communication network. The client-server relationship is created by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system, and solves the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services.

[0107] It should be understood that various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps recited in this application can be executed in parallel, sequentially, or in a different order, as long as the information desired by the technical solution of this application can be achieved, and no limitation is made herein.

[0108] The above specific embodiments do not constitute a limitation on the protection scope of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application shall be included within the protection scope of this application.

Claims

1. A lane recognition method, characterized in that, the method includes: If a target is detected by a radar antenna, determine the detection data of the radar antenna for the target; Determine the lane identifier of the lane where the target is located, and use the detection data and the lane identifier as sample data; Perform model training based on the sample data to determine a lane recognition model, and perform lane recognition based on the lane recognition model.

2. The method according to claim 1, characterized in that, If a target is detected by a radar antenna, determining the detection data of the radar antenna for the target includes: Detect the target distance between the target and the radar antenna through the radar antenna; If the target distance reaches a preset distance, determine the detection data of the radar antenna for the target.

3. The method according to claim 2, characterized in that, If the target distance reaches a preset distance, determining the detection data of the radar antenna for the target includes: If it is detected in the target frame that the target distance reaches the preset distance, determine the detection data of the radar antenna for the target within a preset time traced forward from the target frame.

4. The method according to claim 2, characterized in that, If the preset distance is one, the detection data includes the channel data of the radar antenna; if the preset distance is at least two, the detection data includes the target distance and the channel data.

5. The method according to claim 4, characterized in that, The determination process of the channel data includes: If a target is detected by a radar antenna, determine the target position of the target; Determine the channel data collected by the radar antenna and returned from the target position.

6. The method according to claim 1, characterized in that, The method further includes: If a preset object is detected by a radar antenna, determine the detection data of the radar antenna for the preset object; Input the detection data of the radar antenna for the preset object into the lane recognition model to determine the lane identifier of the lane where the preset object is located.

7. The method according to claim 6, characterized in that, If a preset object is detected by a radar antenna, determining the detection data of the radar antenna for the preset object includes: If it is detected by the radar antenna that the distance of the preset object relative to the radar antenna reaches the preset distance, determine the detection data of the radar antenna for the preset object.

8. A lane recognition device, characterized in that, the device includes: A detection data determination module, configured to determine the detection data of the radar antenna for the target if a target is detected by the radar antenna; A sample data determination module, configured to determine the lane identifier of the lane where the target is located, and use the detection data and the lane identifier as sample data; A model training module, configured to perform model training based on the sample data to determine a lane recognition model, and perform lane recognition based on the lane recognition model.

9. An electronic device, characterized in that, the electronic device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, and when the computer program is executed by the at least one processor, the at least one processor is enabled to perform the lane recognition method according to any one of claims 1-7.

10. A computer-readable storage medium characterized in that the computer-readable storage medium stores computer instructions for causing a processor to perform the lane recognition method according to any one of claims 1-7 when executed.

Images