Method and system for detecting vehicle information

Abstract

The application discloses a kind of detection method and system of vehicle information, wherein the method comprises: in abnormal weather state, the target point cloud data obtained by laser radar according to the first emission power scanning and the reflection intensity of each target point therein are obtained;From the target point cloud data, the target point with the reflection intensity in the target reflection intensity range is selected and / or the non-vehicle point cloud data is removed from the target point cloud data based on the boundary frame, to obtain the target vehicle point cloud data, the target reflection intensity range is the reflection intensity range when scanning to vehicle under the first emission power, the boundary frame is determined according to the reflection intensity of each target point, the last data frame of vehicle point cloud in target vehicle point cloud data or the first data frame only with non-vehicle point cloud thereafter;When target vehicle point cloud data is not empty, the vehicle information of target vehicle matched with target vehicle point cloud data is determined according to target vehicle point cloud data.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle detection, in particular to a vehicle information detection method and system. BACKGROUND

[0002] In order to detect over-limit vehicles, charge vehicles, etc., vehicle profile data of each vehicle passing through a detection area on a road can be collected, and vehicle information such as vehicle width, vehicle length, vehicle model, etc. can be determined based on the collected vehicle profile data, in addition, vehicle flow and other information can also be determined. However, during vehicle driving, water, dust, etc. on the road surface will be brought to the air in the form of water mist or water vapor, dust, etc. when being rolled by the vehicle axle, thereby greatly affecting the collected vehicle profile data, resulting in inaccurate calculation of vehicle length, width, height, etc. information, and even the occurrence of missing vehicles or multiple vehicles.

[0003] Therefore, the vehicle information detection method in the related art has the problem of low accuracy of vehicle information detection due to the influence of interference particles in the air. SUMMARY

[0004] The embodiments of the present application provide a vehicle information detection method and system to at least solve the technical problem of low accuracy of vehicle information detection due to the influence of interference particles in the air in the related art vehicle information detection method.

[0005] According to an aspect of an embodiment of the present application, a vehicle information detection method is provided, comprising: in the case of an abnormal weather state, obtaining target point cloud data obtained by a laser radar scanning a scanning surface of the laser radar according to a first transmission power and a reflection intensity of each target point in the target point cloud data; selecting target points with reflection intensity within a corresponding target reflection intensity range from the target point cloud data and / or removing non-vehicle point cloud data from the target point cloud data based on a boundary frame to obtain target vehicle point cloud data, wherein the target reflection intensity range corresponding to each target point is a reflection intensity range matched with the first transmission power when a vehicle is scanned, and the boundary frame is a last data frame containing vehicle point cloud or a first data frame containing only non-vehicle point cloud after the last data frame containing vehicle point cloud in the target vehicle point cloud data, which is determined according to the reflection intensity of each target point; in the case that the target vehicle point cloud data is not empty, determining vehicle information of a target vehicle matched with the target vehicle point cloud data according to the target vehicle point cloud data.

[0006] According to another aspect of the embodiments of the present application, a vehicle information detection system is also provided, comprising: a laser radar configured to scan a scanning surface of the laser radar at a first emission power in an abnormal weather condition to obtain target point cloud data and a reflection intensity of each target point in the target point cloud data; a data processing component configured to obtain the target point cloud data and the reflection intensity of each target point; select target points with a reflection intensity within a corresponding target reflection intensity range from the target point cloud data and / or remove non-vehicle point cloud data from the target point cloud data based on a boundary frame to obtain target vehicle point cloud data, wherein the target reflection intensity range corresponding to each target point is a reflection intensity range matching the first emission power when a vehicle is scanned, and the boundary frame is determined according to the reflection intensity of each target point, and is the last data frame containing vehicle point cloud or the first data frame containing only non-vehicle point cloud after the last data frame containing vehicle point cloud in the target vehicle point cloud data; and in a case where the target vehicle point cloud data is not empty, determine vehicle information of a target vehicle matching the target vehicle point cloud data according to the target vehicle point cloud data.

[0007] In the embodiment of the present application, the vehicle point cloud information is determined according to the reflection intensity of the point cloud. In the case of abnormal weather conditions, the target point cloud data obtained by the laser radar scanning the scanning surface of the laser radar according to the first transmission power and the reflection intensity of each target point in the target point cloud data are acquired; the target points with the reflection intensity in the corresponding target reflection intensity range are selected from the target point cloud data and / or the non-vehicle point cloud data is removed from the target point cloud data based on the boundary frame to obtain the target vehicle point cloud data, wherein the target reflection intensity range corresponding to each target point is the reflection intensity range matching the first transmission power when the vehicle is scanned, and the boundary frame is the last data frame containing the vehicle point cloud or the first data frame containing only non-vehicle point cloud after the last data frame containing the vehicle point cloud in the target vehicle point cloud data determined according to the reflection intensity of each target point; in the case that the target vehicle point cloud data is not empty, the vehicle information of the target vehicle matching the target vehicle point cloud data is determined according to the target vehicle point cloud data. Since the reflection intensity of a position point when the position point is a vehicle point is different (i.e., not completely the same) from the reflection intensity of the position point when the position point is other points (non-vehicle points, such as other interference particles in water mist or air), the vehicle point cloud data (point cloud data satisfying the reflection intensity condition) is selected from the point cloud data obtained by the laser radar based on the reflection intensity of each point in each point cloud data, so that the information of the corresponding vehicle is determined based on the selected vehicle point cloud data. Since the point cloud data is screened based on the reflection intensity, and the boundary between the data frame containing the vehicle point cloud and the data frame not containing the vehicle point cloud is determined to remove the non-vehicle point cloud data, only the point cloud data containing the interference particles can be removed, and the purpose of reducing the influence of the interference particles in the air on the vehicle information detection can be achieved, and the technical effect of improving the accuracy of vehicle information detection is achieved, thereby solving the technical problem of low accuracy of vehicle information detection in the related art due to the influence of interference particles in the air. BRIEF DESCRIPTION OF DRAWINGS

[0008] The accompanying drawings, which are incorporated into and form a part of the specification, illustrate one embodiment consistent with the present application and, together with the description, serve to explain the principles of the application.

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings needed to be used in the embodiments or prior art description will be briefly introduced as follows. Obviously, those skilled in the art can obtain other drawings from these drawings without any creative effort.

[0010] Figure 1is a schematic diagram of a hardware environment of an optional vehicle information detection method according to an embodiment of the application;

[0011] Figure 2 is a flowchart of an optional vehicle information detection method according to an embodiment of the application;

[0012] Figure 3 is a schematic diagram of an optional vehicle information detection method according to an embodiment of the application;

[0013] Figure 4 is a structural block diagram of an optional electronic device according to an embodiment of the application. DETAILED DESCRIPTION

[0014] In order to enable persons skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the 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 in the present application, all other embodiments obtained by persons skilled in the art without creative work should fall within the scope of protection of the present application.

[0015] It should be noted that the terms "first", "second", and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or a chronological sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" 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 include only those steps or units clearly listed, but can include other steps or units that are not clearly listed or inherent to the process, method, product or device.

[0016] According to an aspect of the embodiments of the present application, a vehicle information detection method is provided. Optionally, in the present embodiment, the vehicle information detection method described above can be applied in a hardware environment containing a vehicle contour identification system 102 and a server 104 as shown in Figure 1 Figure 1 ​As shown, the server 104 is connected with the vehicle contour identification system 102 through a network, can be used to provide services (such as application services, etc.) for the vehicle contour identification system or the client installed on the vehicle contour identification system, and a database can be set on the server or independently of the server to provide data storage services for the server 104. The vehicle contour identification system 102 includes a detection component, which can be but is not limited to a laser sensor, for example, a laser ranging sensor.

[0017] The vehicle information detection method of the embodiment of the application can be executed by the server 104, or by the vehicle contour identification system 102, or by the server 104 and the vehicle contour identification system 102 together.

[0018] Hereinafter, the vehicle information detection method in the embodiment executed by the server 104 is taken as an example for description. Figure 2 is a flowchart of an optional vehicle information detection method according to the embodiment of the application, as shown in the figure, the flow of the method can include the following steps: Figure 2

[0019] In step S202, in the case of abnormal weather state, target point cloud data obtained by the laser radar scanning the scanning surface of the laser radar according to the first transmission power and the reflection intensity of each target point in the target point cloud data are acquired.

[0020] The vehicle information detection method in the embodiment can be applied to the scene of detecting the vehicle information of the passing vehicle in the preset area by the laser radar in the case of abnormal weather state. The abnormal weather can be rainy and snowy weather, weather with a large amount of dust, etc., the above-mentioned preset area can be a highway or other area where vehicle information needs to be detected, and the laser radar can be a perception sensor for scanning vehicle information, which can be set at a specific position of the preset area, for example, the front side of the weighing area, the rear side of the weighing area, etc. The vehicle information here can include the contour information of the vehicle, and can include but is not limited to at least one of the following: vehicle length, vehicle width, vehicle height, etc., and can also include vehicle model information and other vehicle information.

[0021] When using the laser radar to detect the vehicle, due to the high ranging accuracy of the laser radar and the less influence of the ambient light, etc., the accuracy of collecting the vehicle contour data and judging the vehicle type when detecting the vehicle is relatively high. However, in some abnormal weather, the water, dust, etc. on the road surface will be brought into the air by the rolling of the vehicle axle, forming water mist or water vapor, dust particles, and rain, snow, dust, etc. falling in the air, and these particles will reflect the scanning laser, thereby greatly affecting the collected vehicle data.

[0022] ​In the related art, in order to improve the accuracy of vehicle detection in abnormal weather, a multi-sensor fusion technology is usually used, for example, laser radar, video, millimeter wave are combined, and a deep learning method can also be used to process rain and fog data detected by laser radar: convert laser radar point cloud data into point cloud picture data, use a deep learning method to improve vehicle recognition accuracy, for example, use point cloud density distribution to judge the weather. However, multi-sensor fusion will increase the cost of vehicle detection, and the use of deep learning methods will increase the complexity of vehicle detection.

[0023] In this embodiment, in the case of abnormal weather state, the vehicle can be scanned by the laser radar, and the scanned point cloud data may contain point cloud data other than vehicle points (i.e. non-vehicle points, or non-vehicle point cloud data). For example, when the scanned vehicle point cloud data contains non-vehicle points such as water mist and dust, the scanned point cloud data can be filtered to determine the possible vehicle point cloud data, and then the vehicle information is calculated, thereby avoiding the interference of the interference particles in the air on the vehicle information detection. In some examples of this embodiment, the interference particles are water mist, and the expected filtered point cloud data is water mist point cloud data.

[0024] In this embodiment, the laser radar can scan the scanning surface of the laser radar according to the first transmission power, thereby obtaining the target point cloud data. The laser radar used can be a horizontal scanning laser radar for scanning the cross section of the vehicle. The vehicle scanning performed by the laser radar can be continuously performed, and the trigger condition for performing vehicle scanning can be to determine that a vehicle passes, or it can also be to continuously perform vehicle scanning according to a predetermined frequency regardless of whether a vehicle passes. In addition, while obtaining the target point cloud data, the reflection intensity of each target point in the target point cloud data can also be obtained.

[0025] The transmission frequency allowed to be used by the laser radar for vehicle scanning can be a transmission frequency in the transmission power range. The transmission power range can be a set of discrete transmission frequencies or a set of transmission frequencies from the minimum transmission frequency to the maximum transmission frequency. The first transmission frequency can be any one of the transmission powers in the above-mentioned transmission power range. The target point cloud data scanned by the laser radar can be sent to a server (or other processing device), and the server (which can also be other data processors or data processing components) can obtain the above-mentioned target vehicle data.

[0026] The target point cloud data can be point cloud data scanned by the lidar at the first transmission power. Each target point in the target point cloud data corresponds to a scanning point in the scanning surface. The target point cloud data can or can not contain vehicle points; and can or can not contain non-vehicle points (e.g., points corresponding to interference particles such as dust, water mist, etc.). The target vehicle point cloud data herein can be point cloud data obtained by the lidar by performing multiple vehicle scans. Alternatively, the target point cloud data is a complete point cloud data integrated from multiple frames of scanned vehicle point cloud data, for example, point cloud data integrated from multiple frames of point cloud data from a vehicle start frame to a vehicle end frame.

[0027] It should be noted that in an abnormal weather state, a large number of interference particles can appear in the air, for example, in rainy and snowy weather, the corresponding interference particles include water mist, and for example, in dusty weather, the corresponding interference particles include dust. The interference particles herein can be an interference particle region caused by vehicle movement. The vehicle information detection method in the embodiment can be performed in the case of detecting an abnormal weather state.

[0028] For example, the rainy and snowy weather can be detected and determined, so that the water mist point cloud data scanned in the vehicle point cloud data is effectively removed based on the preset reflection intensity range, the interference of the water mist on the vehicle information calculation is reduced, and the accuracy of vehicle detection is improved.

[0029] In step S204, target points with reflection intensity within the corresponding target reflection intensity range are selected from the target point cloud data, and / or non-vehicle point cloud data is removed from the target point cloud data based on the boundary frame, to obtain target vehicle point cloud data.

[0030] Under the same transmission power, the reflection intensities of the vehicle and the non-vehicle are different at the same scanning point in the scanning surface, and under different transmission powers, the reflection intensities of the vehicle are different at the same scanning point (scanning points with the same distance and the same angle) in the scanning surface, and the reflection intensities of the non-vehicle will also be different. Therefore, the reflection intensity range of each scanning point in the scanning surface corresponding to the lidar when the scanning point is a vehicle point can be pre-configured to match each transmission power, and the reflection intensity range of the scanning point in the scanning surface corresponding to the lidar when the scanning point is a non-vehicle point can also be pre-configured to match each transmission power.

[0031] In the embodiment, for the target point cloud data, the scanning points corresponding to each target point in the target point cloud data can be determined respectively, and the target reflection intensity range corresponding to each target point can be determined, where the target reflection intensity range corresponding to each target point is the reflection intensity range when the vehicle is scanned (i.e., the reflection intensity range when the scanning point corresponding to the target point is the vehicle point) matching the first transmission power; the target points with the reflection intensity in the corresponding target reflection intensity range are selected from the target point cloud data, so as to obtain the target vehicle point cloud data, and the eliminated points can be considered as non-vehicle points, such as interference particle points.

[0032] The selection of the target points described above can be performed all the time, that is, the selection of the target points described above is performed regardless of the weather state, or can be performed after it is determined that there is interference particle (such as water mist), or can be performed after it is determined that the current weather is abnormal weather (such as rainy and snowy weather), which is not limited in the embodiment.

[0033] For example, taking the interference particle as water mist and the corresponding weather state as rainy and snowy weather as an example, the set of transmission powers of the laser radar is: [P1, P2, …, Pn], where n is a positive integer greater than 1. In the rainy and snowy weather, the laser radar detects the reflection intensity value of each point (i.e., target point) in the first vehicle point cloud data (i.e., target point cloud data) at the transmission power P1, P2, …, Pn (i.e., P1, P2, …, Pn) respectively, and the reflection intensity value of each point in the first reflection intensity set is composed of the reflection intensity value detected at the transmission power P1, P2, …, Pn (i.e., P1, P2, …, Pn). n ], wherein n is a positive integer greater than 1. In the rainy and snowy weather, the laser radar detects the reflection intensity value of each point (i.e., target point) in the first vehicle point cloud data (i.e., target point cloud data) at the transmission power P i (i∈[1,n]), and the reflection intensity value of each point in the first reflection intensity set is composed of the reflection intensity value detected at the transmission power P il (i∈[1,n]). (i.e., the reflection intensity range of each scanning point in the scanning surface under the condition that the transmission power is P i (i∈[1,n]), and the reflection intensity value of each point in the second reflection intensity set is a range value. Comparing the first reflection intensity set and the second reflection intensity set, if the reflection intensity value of the corresponding point in the first reflection intensity set is not in the reflection intensity range of the corresponding point in the second reflection intensity set, it is judged that the point is a water mist point, and the point data is eliminated in the vehicle point cloud data, and the remaining vehicle point cloud data is the second vehicle point cloud data (i.e., target vehicle point cloud data).

[0034] Here, considering different categories of vehicles (for example, large vehicles and small vehicles), the reflection intensity value of the same position point under the same transmission power will also be different, and the reflection intensity range that can be configured for each scanning point under each transmission power can be multiple, corresponding to different vehicle categories. Taking the rough classification of vehicles into two categories of small vehicles and large vehicles as an example, the rough classification of vehicles can be performed according to the vehicle point cloud data, and the fourth reflection intensity range D i1The same point cloud data of the coarse classification vehicle is selected to form a second reflection intensity set. In addition, the reflection intensity of the water mist point can be counted into the reflection intensity range of the laser radar scanning to the water mist, that is, the reflection intensity range when the scanning point corresponding to the transmission power P i When the scanning point corresponding to the transmission power P

[0035] In this embodiment, considering that the reflection intensity values of the same position points under the same transmission power are different for different categories of vehicles, in order to improve the accuracy of vehicle information identification, a set of vehicle categories can be pre-configured, which can include multiple vehicle categories, such as large vehicles and small vehicles. Before selecting the target points with reflection intensity in the corresponding target reflection intensity range from the target point cloud data, the expected vehicle category in the set of vehicle categories that matches the target point cloud data can be determined according to the point cloud characteristics of the target point cloud data; the target reflection intensity range corresponding to each target point can be determined according to the reflection intensity range when each scanning point in the scanning surface is a vehicle point under the first transmission power and the expected vehicle category.

[0036] Optionally, the following information can be pre-configured for the vehicles in each vehicle category in the set of vehicle categories: the reflection intensity range when each scanning point is a vehicle point under different transmission powers and different vehicle categories, and the reflection intensity range when each scanning point is a non-vehicle point under different transmission powers and different vehicle categories. The configured reflection intensity range can include the reflection intensity range corresponding to the abnormal weather state, and can also include the reflection intensity range corresponding to the non-abnormal weather state.

[0037] For the target point cloud data, the expected vehicle category in the set of vehicle categories that matches the target point cloud data can be determined according to the point cloud characteristics of the target point cloud data, and the point cloud characteristics can be used to determine the vehicle characteristics of the vehicle that matches the target point cloud data. The vehicle characteristics can include at least one of the following characteristics: length characteristics, width characteristics, and height characteristics. The vehicle category here can be considered as a kind of coarse classification category, and the actual vehicle category can be calculated when determining the vehicle information; the expected vehicle category can include but is not limited to small vehicles and large vehicles.

[0038] According to the determined expected vehicle category, the reflection intensity range when each scanning point in the scanning surface is a vehicle point under the first transmission power and the expected vehicle category can be determined, so as to obtain the target reflection intensity range corresponding to each target point. The target reflection intensity range corresponding to each target point can be the target reflection intensity range corresponding to each target point under the preset weather state.

[0039] Optionally, due to a large number of interference particles (for example, water mist) in the air under abnormal weather conditions, some frame point cloud data in the target point cloud data is all non-vehicle point cloud data, resulting in low accuracy of the determined vehicle information. In this case, the boundary frame in the target point cloud data can be determined, and the non-vehicle point cloud data is removed from the target point cloud data based on the boundary frame to obtain target vehicle point cloud data. Here, the boundary frame is the last frame containing vehicle point cloud or the first frame containing only non-vehicle point cloud after the last frame containing vehicle point cloud in the multiple frame point cloud data determined according to the reflection intensity of each target point. According to the determined boundary frame, the point cloud data after the boundary frame in the target point cloud data or the boundary frame and the point cloud data after the boundary frame can be removed, and the target vehicle point cloud data is obtained. After the above operation, the obtained target vehicle point cloud data can be a continuous multiple frame point cloud data containing only vehicle point cloud data.

[0040] For example, in the second vehicle point cloud data, the boundary frame in the multiple frame point cloud data can be determined from the first frame of the vehicle, that is, the vehicle and water mist boundary frame, for example, the jth frame, and the jth frame to the end frame of the vehicle are all water mist.

[0041] Optionally, for the determination of the target vehicle point cloud data, any one of the above methods of selecting target points and removing non-vehicle point cloud data based on the boundary frame can be used, or both methods can be used together.

[0042] In step S206, if the target vehicle point cloud data is not empty, the vehicle information of the target vehicle matching the target vehicle point cloud data is determined according to the target vehicle point cloud data.

[0043] If the non-vehicle points in the target point cloud data are removed and the target vehicle point cloud data is empty, it can be determined that the target point cloud data is all non-vehicle point cloud data, and the subsequent vehicle information processing operation is not performed. If the target vehicle point cloud data is not empty, it indicates that the target vehicle point cloud data can be vehicle point cloud data of a certain vehicle (i.e., the target vehicle), and the vehicle information processing operation can be performed, that is, the vehicle information of the target vehicle matching the target vehicle point cloud data is determined according to the target vehicle point cloud data. The categories and determination methods of the vehicle information are similar to the above, and are not described here.

[0044] For example, if the water mist points are removed and there is no second vehicle point cloud data, it is judged that the detected point cloud data is all water mist point cloud data, and the vehicle information is not calculated. If the water mist points are removed and there is second vehicle point cloud data, the vehicle information of the target vehicle matching the second vehicle point cloud data can be determined based on the second vehicle point cloud data.

[0045] After obtaining the target vehicle point cloud data, if the target vehicle point cloud data is not empty, vehicle information detection can be directly performed based on the target vehicle point cloud data to obtain vehicle information of the target vehicle matched with the target vehicle point cloud data. However, the reflection intensity range of the same scanning point when it is a vehicle point and the reflection intensity range when it is a non-vehicle point under the same transmission power are not completely non-overlapping, that is, there is an intersection between the two, and therefore, the target vehicle point cloud data can contain non-vehicle point cloud data. In order to improve the accuracy of vehicle detection, the target vehicle point cloud data can be further screened and updated, and the above screening can be screening based on the vehicle range.

[0046] Optionally, after selecting the target points with the reflection intensity within the corresponding target reflection intensity range from the target point cloud data to obtain the target vehicle point cloud data, in the case that the target vehicle point cloud data is not empty, the target vehicle point cloud data is converted into a first vehicle point cloud image with the vehicle length direction as the main display direction and a second vehicle point cloud image with the vehicle width direction as the main display direction, respectively; the first vehicle point cloud image and the second vehicle point cloud image are identified to determine the target vehicle range of the target vehicle; according to the corresponding relationship between the target vehicle point cloud data and the image pixels in the first vehicle point cloud image and the second vehicle point cloud image, the target points in the target vehicle point cloud data outside the target vehicle range are removed to obtain updated target vehicle point cloud data, and the removed target points can be considered as non-vehicle point cloud data, that is, interference point cloud data.

[0047] Here, the conversion process of the point cloud data to the vehicle point cloud image can be performed by a point cloud data conversion module. The point cloud data conversion module can first determine the vehicle length direction and the vehicle width direction based on the point cloud features or based on the scanning direction of the laser radar, and then project the target vehicle point cloud data to the first reference plane corresponding to the vehicle length direction and the second reference plane corresponding to the vehicle width direction to obtain the first vehicle point cloud image and the second vehicle point cloud image. The identification of the vehicle point cloud image can be performed by a point cloud image identification module. The point cloud image identification module can input the first vehicle point cloud image and the second vehicle point cloud image into a pre-trained image recognition model and obtain the vehicle range output by the image recognition model to obtain the target vehicle range.

[0048] For example, the second vehicle point cloud data can be converted into a vehicle point cloud image displayed mainly in the vehicle length direction on the front of the image and a vehicle point cloud image displayed mainly in the vehicle width direction on the front of the image; the vehicle point cloud images are identified using a deep learning method, and the vehicle range is selected from the two vehicle point cloud images; according to the correspondence between the second vehicle point cloud data and the image pixels, the fog point cloud in the second vehicle point cloud data is selected and the fog point cloud is removed. In addition, the reflection intensity of the fog point can also be counted into the reflection intensity range of the laser radar scanning to the fog in a similar manner as described above.

[0049] Here, by converting the vehicle point cloud data into vehicle point cloud images in the vehicle length direction and the vehicle width direction respectively, and then identifying the vehicle range, the vehicle point cloud data is screened again based on the identified vehicle range, which can improve the accuracy of vehicle information detection. For rainy and snowy weather, by converting the point cloud data into point cloud picture data and using a deep learning method to identify vehicles, the interference of fog point cloud data on vehicles can be reduced, thereby improving the accuracy of laser radar in detecting vehicles in rainy and snowy weather.

[0050] Through the above steps S202 to S206, the target point cloud data obtained by the laser radar scanning the scanning surface according to the first emission power and the reflection intensity of each target point in the target point cloud data are obtained; the target points with the reflection intensity within the corresponding target reflection intensity range are selected from the target point cloud data and / or the non-vehicle point cloud data is removed from the target point cloud data based on the boundary frame to obtain the target vehicle point cloud data, wherein the target reflection intensity range corresponding to each target point is the reflection intensity range matched with the first emission power when the vehicle is scanned, and the boundary frame is the last data frame containing the vehicle point cloud or the first data frame containing only non-vehicle point cloud after the last data frame containing the vehicle point cloud in the target vehicle point cloud data determined according to the reflection intensity of each target point; in the case that the target vehicle point cloud data is not empty, the vehicle information of the target vehicle matched with the target vehicle point cloud data is determined according to the target vehicle point cloud data, which solves the technical problem of low accuracy of vehicle information detection in the related art due to the influence of interference particles in the air, and improves the accuracy of vehicle information detection.

[0051] In one example embodiment, the above method further comprises:

[0052] S11, determining a reference intensity value of each scanning point in the scanning surface, wherein the reference intensity value of each scanning point is the average value of the reflection intensity range matched with the first emission power when each scanning point is a vehicle point;

[0053] S12, cyclically performing the boundary frame determination step until a loop end condition is met, wherein the loop end condition comprises that N is 2 or the boundary frame is determined.

[0054] In the embodiment, the boundary frame in the target point cloud data can be determined according to the reference intensity value of each scanning point in the scanning plane, the target point cloud data can contain multiple frames of point cloud data, and the boundary frame can be determined from the multiple frames of point cloud data. In this regard, the reference intensity value of each scanning point can be first determined, and the reference intensity value of each scanning point here can be the average of the reflection intensity range matching the first transmission power when each scanning point is a vehicle point. After determining the reference intensity value of each scanning point, the boundary frame can be determined by performing the following boundary frame determination step. The boundary frame determination step can be cyclically performed until a loop end condition is met. The loop end condition here can include one of the following: N is 2, the boundary frame is determined, and the boundary frame has been determined after performing the boundary frame determination step in this round.

[0055] In the process of performing the following steps once, N (N > 2) consecutive frames can be first selected from the multiple frames of point cloud data contained in the target point cloud data in sequence, and the N consecutive frames are taken as a frame data detection block. In the N consecutive frames, the missing points of each frame in the N consecutive frames can be determined, where N is a positive integer greater than 2; the missing points of each frame can be target points contained in at least one frame other than each frame in the N consecutive frames and not in each frame. According to the determined missing points of each frame, the reference intensity value of each scanning point can be used to expand the missing points of each frame to obtain an expanded each frame. The points contained in the expanded each frame can be the same, and the value of the missing points of each frame is the reference intensity value of the corresponding scanning point.

[0056] For example, for the current frame, if one of the N consecutive frames has reflection intensity at the scanning point (8, 7) (i.e., there is point cloud data), but the current frame does not have reflection intensity at the scanning point, the scanning point can be considered as a missing point of the current frame, and the average of the reflection intensity range when the scanning point is a vehicle point under the current transmission power is configured as the reflection intensity of the scanning point of the current frame.

[0057] It should be noted that for the process of determining the missing points of each frame in the N consecutive frames, the following process can be used: first, determine the union of the target points contained in the N consecutive frames to obtain a reference point set, and multiple target points corresponding to the same scanning point in the N consecutive frames are merged into one; determine the missing points in the current frame, and the missing points in the current frame are target points contained in the reference point set but not in the current frame.

[0058] Each frame obtained contains the same points, but different frames contain different reflection intensities of the points matched with the same scanning point. In this regard, the mean value of the reflection intensities of the corresponding points in each expanded frame (i.e., the mean value of the points matched with the same scanning point in each frame, respectively) can be calculated and determined as the reflection intensity of the point matched with the scanning point in the reflection intensity mean value frame, thereby obtaining the reflection intensity mean value frame. The reflection intensity mean value frame contains the same points as each expanded frame, i.e., the same number of points and the same corresponding scanning points.

[0059] For each frame, the variance of the reflection intensities of the points in each frame before expansion and the reflection intensities of the corresponding points in the reflection intensity mean value frame can be determined to obtain the reflection intensity variance corresponding to each frame. The reflection intensity variances corresponding to different frames can be the same or different.

[0060] If there is a first frame in the N consecutive frames that satisfies the following conditions, the first frame can be determined as a boundary frame: the reflection intensity variance corresponding to the first frame is greater than or equal to a first variance threshold; the variances corresponding to the first frame to the last frame in the frame point cloud data are all greater than or equal to the first variance threshold, i.e., the reflection intensity variances corresponding to the first frame and each frame after the first frame are all greater than or equal to the first variance threshold; and the variance of the corresponding points of the first frame and the frame before the first frame is greater than or equal to a second variance threshold.

[0061] If no boundary frame is determined until the last frame data detection block of the multi-frame point cloud data, it can be determined that no boundary frame is determined from the multi-frame point cloud data, and the value of N obtained after N is reduced by 1 can be used to update N to obtain an updated N. The first variance threshold can be a value that is pre-set and can be adjusted according to actual conditions. The second variance threshold is similar to the first variance threshold and will not be described here.

[0062] For example, in the second vehicle point cloud data, starting from the first frame of the vehicle (the vehicle start frame), take N (N > 2) consecutive frames as a frame data detection block, use the average value of the reflection intensity range of the corresponding points of the vehicle coarse classification to expand the reflection intensity value of the corresponding points in the N frames, that is, compare one frame with other frames, if there is no corresponding point, use the average value of the reflection intensity range of the corresponding points of the vehicle coarse classification to expand, so that the number of corresponding points in each frame is consistent, calculate the average value of the corresponding points in the N frames to obtain the reflection intensity average value frame. Then, the reflection intensity variance of the corresponding points of each frame in the N frames without expansion and the reflection intensity average value frame is calculated in turn. When the reflection intensity variance of the jth frame is greater than the first variance threshold, and the variances of the jth frame to the vehicle end frame are all greater than the first variance threshold, and the variance of the corresponding points of the jth frame and the j-1th frame is greater than the second variance threshold, then the jth frame to the vehicle end frame are all water mist; when N consecutive frames are not calculated, take N-1 consecutive frames to calculate, until the vehicle and water mist boundary frame are calculated, or until 2 consecutive frames are taken.

[0063] Through the embodiment, by expanding different frames, determining the average value frame based on the reflection intensity of the expanded multiple frames, and determining the boundary frame based on the variance of the reflection intensity of each expanded frame and the reflection intensity of the average value frame, the accuracy of the boundary frame determination can be improved.

[0064] In one example embodiment, the abnormal weather state can be a rainy and snowy weather state. Under the rainy and snowy weather, water mist can be brought up in the air due to vehicle driving and other reasons, and the water mist can be scanned by the laser radar, thereby affecting the accuracy of vehicle information detection.

[0065] For example, the main reason why the rainy and snowy weather affects the use of the laser radar is that the data directly scanned by the laser radar when the raindrops or snowflakes are processed by using the filtering method, so that the point cloud data of the scanned vehicle is obtained. However, when the vehicle passes through the range of the laser radar, the water mist formed by the water on the road surface is adhered to the upper part of the road surface, and a water mist surface or a water mist belt is formed above the road surface. The laser radar scans the vehicle, and also scans the water mist surface or the water mist belt, thereby affecting the detection of the vehicle information by the laser radar. For example, the water mist surface or the water mist belt is misidentified as part of the vehicle, thereby reducing the accuracy of the vehicle information detection.

[0066] Correspondingly, in order to determine whether the current is in the rainy and snowy weather state, the above method further includes:

[0067] S21, in the case that the point cloud data of the K consecutive vehicles all exist water mist point cloud data, it is determined that the current is in the rainy and snowy weather state, wherein K is a positive integer greater than or equal to 2;

[0068] S22. When it is determined that there is no water mist point cloud data in the point cloud data of M consecutive vehicles, or the reflection intensity variance of each road surface point in the scanning surface in T consecutive frames is greater than or equal to the third difference threshold, it is determined that the current weather state is not rainy or snowy, where M and T are both positive integers greater than or equal to 2.

[0069] In this embodiment, the current weather conditions can be determined using point cloud data obtained from LiDAR scanning. If water mist point cloud data is determined to be present in the point cloud data of K consecutive vehicles, it can be determined that the current weather condition is rainy or snowy. Determining the presence of water mist point cloud data in the point cloud data of a vehicle can be accomplished by determining that multiple consecutive frames of the vehicle's point cloud data contain only water mist point clouds, or by other methods.

[0070] If it is determined that no water mist point cloud data exists in the point cloud data of M consecutive vehicles, it can be determined that the current weather condition is not rainy or snowy. Determining that no water mist point cloud data exists in the point cloud data of a vehicle can be done by determining that the vehicle's point cloud data does not contain multiple consecutive frames containing only water mist point clouds. Here, K can be a positive integer greater than or equal to 2, M and T can both be positive integers greater than or equal to 2, and M and K can be the same or different.

[0071] Alternatively, because reflected light is generally stable (the reflection intensity at different ground points is essentially the same) with a small variance when the ground is wet or snowy, and because emitted light is generally unstable (the reflection intensity at different ground points varies more significantly) due to unevenness and other factors, the current weather conditions can be determined based on the variance of the reflection intensity at each road surface point within the scanned surface of each frame in the vehicle's point cloud data. A third-party difference threshold can be pre-set; if the variance of the reflection intensity at each road surface point within the scanned surface in T consecutive frames is greater than or equal to the third-party difference threshold, it can be determined that the current weather condition is not rainy or snowy.

[0072] For example, the laser radar scans objects within the detection range, takes points within the height threshold range to form the first point cloud data, and takes the value of each point in the world coordinate system (f, n, p, x, y, z), where f represents the frame index of the laser radar scanning vehicle, n represents the point index in the frame, p represents the reflection intensity of the point, x represents the length, which is parallel to the vehicle's driving direction, y represents the bandwidth, which is perpendicular to the vehicle's driving direction, and z represents the height. Starting from the vehicle's first threshold index frame to the vehicle's end frame, it is determined whether water mist is detected from the point cloud data of each vehicle. When water mist is detected for K consecutive vehicles, it is determined that the current weather is in rainy and snowy conditions; when water mist is not detected for M consecutive vehicles, and / or the reflection emphasis variance of all points on the road surface in T consecutive frames scanned by the laser radar within the detection road range is greater than the third difference threshold, it is determined that the current weather is in a non-rainy and non-snowy condition.

[0073] It should be noted that the target point cloud data or the point cloud data of any vehicle can be point cloud data formed by points within a height threshold range scanned by a laser radar. The data of each point in the point cloud data includes at least one of the following: the position of each point in a world coordinate system, the reflection intensity of each point, the index of the frame scanned by the laser radar where each point is located (i.e., the frame index corresponding to each point), the index of each point in the frame where each point is located (the point index of each point in the frame), and the like, and can also include other information, which is not limited in the present embodiment.

[0074] According to the present embodiment, whether the current is in a rainy and snowy weather state is determined according to the scanned point cloud data, which can improve the accuracy and convenience of weather state determination, thereby improving the accuracy of vehicle information detection.

[0075] In one example embodiment, the above method further comprises:

[0076] S31, determining a vehicle start frame and a vehicle end frame corresponding to the detected current vehicle;

[0077] S32, determining a vehicle profile parameter corresponding to each frame from the vehicle start frame to the vehicle end frame, wherein the vehicle profile parameter includes at least one of the following: a vehicle height, a vehicle width; in the case that the vehicle profile parameters corresponding to the continuous Q frames are all less than or equal to a preset parameter threshold, it is determined that there are continuous multiple frames containing non-vehicle point cloud data in the point cloud data of the current vehicle, wherein Q is a positive integer greater than or equal to 2; or,

[0078] S33, determining a reflection intensity change variance corresponding to each frame from the vehicle start frame to the vehicle end frame, wherein the reflection intensity change variance corresponding to each frame is the variance of the reflection intensity of the points in each frame and the corresponding points in the previous frame of each frame; in the case that there is a second frame corresponding to a reflection intensity change variance greater than or equal to a fourth variance threshold in the vehicle start frame to the vehicle end frame, and the reflection intensity change variances corresponding to the continuous L frames starting from the second frame are less than or equal to a fifth variance threshold, and the variance of the reflection intensity of the points in the continuous L frames and the corresponding points in the previous frame of the second frame is greater than or equal to the fourth variance threshold, it is determined that there are continuous multiple frames containing only non-vehicle points in the point cloud data of the current vehicle, wherein L is a positive integer greater than or equal to 2.

[0079] In the embodiment, in order to determine whether the point cloud data of a vehicle contains water mist point cloud data, a vehicle start frame and a vehicle end frame corresponding to the detected current vehicle can be determined first, and the operation of detecting the vehicle start frame and the vehicle end frame can be performed by a vehicle detection module. The detection process can refer to related technologies (such as vehicle separation technology), which are not limited in the embodiment.

[0080] As an optional implementation, for each frame from the vehicle start frame to the vehicle end frame, a corresponding vehicle contour parameter can be determined, which is used to describe the contour of the vehicle and can include at least one of the following: vehicle height, vehicle width. If the vehicle contour parameters corresponding to the continuous Q frames are all less than or equal to a preset parameter threshold, it can be determined that the point cloud data of the current vehicle contains non-vehicle point cloud data in the continuous multiple frames. Here, Q is a positive integer greater than or equal to 2. The preset parameter threshold can be a threshold of the vehicle contour parameter, which can include but is not limited to at least one of the following: a vehicle height threshold, a vehicle width threshold.

[0081] As another optional implementation, the reflection intensity variance corresponding to each frame from the vehicle start frame to the vehicle end frame can also be determined, that is, the variance of the reflection intensity of the points in each frame and the corresponding points in the previous frame of each frame. Then, if it is determined that there is a second frame corresponding to the reflection intensity variance greater than or equal to a fourth variance threshold from the vehicle start frame to the vehicle end frame, the reflection intensity variances corresponding to the continuous L frames starting from the second frame are less than or equal to a fifth variance threshold, and the variance of the reflection intensity of the points in the continuous L frames and the corresponding points in the previous frame of the second frame is greater than or equal to the fourth variance threshold, it can be determined that the point cloud data of the current vehicle contains water mist point cloud data. L is a positive integer greater than or equal to 2.

[0082] For example, the height and width of each frame of the vehicle are calculated from the vehicle first threshold index frame (i.e., the vehicle start frame) to the vehicle end frame. When the height of the continuous M frames in the vehicle is less than the second height threshold and / or the width is less than the first width threshold, it is determined that there is water mist in the vehicle. And / or, the reflection intensity variance (or reflection intensity variance) of all corresponding points (i.e., points with the same scanning height and angle) in adjacent frames can also be calculated. When the reflection intensity variance is greater than the first variance threshold, and the adjacent reflection intensity variances of the continuous L frames starting from the frame are all less than the fifth variance threshold, and the reflection intensity variance of the previous frame of the frame is greater than the low vulgar variance threshold, it is determined that the vehicle contains water mist.

[0083] By the embodiment, based on the vehicle contour parameters corresponding to each frame in the point cloud data and / or the reflection intensity variance of the same points in adjacent frames, it is determined whether there is continuous multi-frame containing water mist point cloud data in the vehicle point cloud data, so that the flexibility and accuracy of information detection can be improved.

[0084] In one example embodiment, the first transmission power is one of a set of transmission powers, which can include multiple transmission powers, for example, [P1, P2,..., P n Correspondingly, the method further includes:

[0085] S41, selecting a second transmission power from a set of transmission powers according to a first reflection intensity range corresponding to each transmission power in the set of transmission powers and a second reflection intensity range corresponding to each transmission power, wherein the first reflection intensity range corresponding to each transmission power is a reflection intensity range matched with each transmission power when the scanning surface is to the vehicle, and the second reflection intensity range corresponding to each transmission power is a reflection intensity range matched with each transmission power when the scanning is to a non-vehicle point;

[0086] S42, adjusting the transmission power of the laser radar from the first transmission power to the second transmission power.

[0087] The reflection intensity range when the same scanning point is a vehicle point and the reflection intensity range when the same scanning point is a non-vehicle point are not exactly the same, and the overlap between the two is also not the same under different transmission powers. When the overlap is less, the possibility of non-vehicle points contained in the selected vehicle point cloud data is smaller; on the contrary, when the overlap is more, the possibility of non-vehicle points contained in the selected vehicle point cloud data is greater. Therefore, under different transmission powers, there will also be differences in the accuracy of vehicle information detection.

[0088] In order to improve the accuracy of vehicle information detection, the transmission power of the scanning detection component can be adjusted based on the detected information. The above transmission power adjustment operation can be determined based on the following information: the reflection intensity range when the scanning surface is to the vehicle under each transmission power, i.e. the first reflection intensity range corresponding to each transmission power; the reflection intensity range when the scanning is to a non-vehicle point under each transmission power, i.e. the second reflection intensity range corresponding to each transmission power. The above transmission power adjustment operation can be performed by a transmission power adjustment module.

[0089] According to the first reflection intensity range corresponding to each transmission power and the second reflection intensity range corresponding to each transmission power, the second transmission power can be selected from the set of transmission powers. After the second transmission power is selected, the transmission power of the laser radar can be adjusted from the first transmission power to the second transmission power, so that the second transmission power can be used to detect the subsequent vehicle.

[0090] Alternatively, the second transmission power can be determined according to the intersection of the first reflection intensity range corresponding to each transmission power and the second reflection intensity range corresponding to each transmission power, or the second transmission power can be determined according to the size of the first reflection intensity range corresponding to each transmission power and the second reflection intensity range corresponding to each transmission power, or the second transmission power can be determined according to other information, which is not limited in the embodiment.

[0091] Through the embodiment, by dynamically adjusting the transmission power of the laser radar, the laser radar can be ensured to use a suitable transmission power for vehicle detection, thereby improving the accuracy of vehicle information detection.

[0092] In an example embodiment, according to the first reflection intensity range corresponding to each transmission power and the second reflection intensity range corresponding to each transmission power, the second transmission power is selected from the set of transmission powers, comprising:

[0093] S51, according to the range intersection between the first reflection intensity range corresponding to each transmission power and the second reflection intensity range corresponding to each transmission power, and the range size of the first reflection intensity range corresponding to each transmission power, the second transmission power is selected from the set of transmission powers.

[0094] In the embodiment, the second transmission power can be selected from the set of transmission powers according to one or more selection parameters corresponding to each transmission power, and the one or more selection parameters can include at least one of the following: the range intersection between the first reflection intensity range corresponding to each transmission power and the second reflection intensity range corresponding to each transmission power, and the range size of the first reflection intensity range corresponding to each transmission power. The selection can be based on small range intersection and moderate first reflection intensity range (too small to easily eliminate vehicle points, too large to easily retain too many non-vehicle points).

[0095] There can be various ways to select the second transmission power from the set of transmission powers, for example, a set of candidate transmission powers can be selected according to one selection parameter, and the second transmission power can be selected according to another selection parameter, and the selection parameter is one of the selection parameters.

[0096] Exemplarily, a range intersection between the first reflection intensity range corresponding to each transmission power and the second reflection intensity range corresponding to each transmission power can be determined first, such as a range intersection between the reflection intensity range matching each transmission power when each scanning point in the scanning surface is a vehicle point and the reflection intensity range matching each transmission power when each scanning point in the scanning surface is a vehicle point, to obtain a range intersection corresponding to each scanning point under each transmission power; an average value, a maximum value or other value of the range intersection corresponding to each scanning point under each transmission power is determined as the range intersection between the first reflection intensity range corresponding to each transmission power and the second reflection intensity range corresponding to each transmission power; then, one or more transmission powers with a range intersection less than or equal to a preset threshold or a preset number of transmission powers with a range intersection at the rear are selected as a group of candidate transmission powers; finally, a transmission power with a maximum first reflection intensity range is selected from the group of candidate transmission powers, to obtain the second transmission power.

[0097] For example, based on the reflection intensity range of the laser radar scanning on the vehicle and the reflection intensity range of the laser radar scanning on the water mist under each transmission power, the transmission power of the laser radar can be adjusted to P k to minimize the intersection of the reflection intensity range of the laser radar scanning on the vehicle and the reflection intensity range of the laser radar scanning on the water mist, and to optimize the reflection intensity range of the laser radar scanning on the vehicle.

[0098] Through the embodiment, the transmission power of the laser radar is dynamically adjusted based on the reflection intensity range of the laser radar scanning on the vehicle and the reflection intensity range of the laser radar scanning on the water mist under each transmission power, which can improve the accuracy of vehicle information detection.

[0099] In one exemplary embodiment, before the second transmission power is selected from the group of transmission powers according to the first reflection intensity range corresponding to each transmission power and the second reflection intensity range corresponding to each transmission power, the above method further comprises:

[0100] S61, adjusting the first reflection intensity range corresponding to the first transmission power and the second reflection intensity range corresponding to the first transmission power according to the reflection intensity of the target point in the target vehicle point cloud data and the reflection intensity of other target points in the target point cloud data except the target vehicle point cloud data;

[0101] S62, determining a reflection intensity correction value according to the first reflection intensity range corresponding to the first transmission power before adjustment and the first reflection intensity range corresponding to the first transmission power after adjustment;

[0102] S63, adjusting the first reflection intensity range corresponding to the other transmission power in the set of transmission powers except the first transmission power using the reflection intensity correction value, to obtain the adjusted first reflection intensity range corresponding to the other transmission power;

[0103] S64, adjusting the second reflection intensity range corresponding to the other transmission power according to the first transmission power, the adjusted second reflection intensity range corresponding to the first transmission power, the other transmission power and the reflection intensity decay rate, to obtain the adjusted second reflection intensity range corresponding to the other transmission power.

[0104] Before selecting the second transmission power from the set of transmission powers according to the first reflection intensity range corresponding to each transmission power and the second reflection intensity range corresponding to each transmission power, the first reflection intensity range and the second reflection intensity range corresponding to each transmission power can be corrected to obtain more accurate first reflection intensity range and second reflection intensity range.

[0105] In order to improve the accuracy of the laser radar transmission power adjustment, the reflection power of the vehicle point and the reflection power of the non-vehicle point detected each time the vehicle information is detected can be used to adjust the first reflection intensity range and the second reflection intensity range corresponding to the first transmission power, and the first reflection intensity range and the second reflection intensity range corresponding to the other transmission power are adjusted based on the adjusted reflection intensity range; and the adjusted transmission power is determined according to the adjusted first reflection intensity range and the second reflection intensity range corresponding to each transmission power. The above reflection intensity calculation and adjustment operations can be performed by the reflection intensity calculation and statistics module.

[0106] For example, the laser radar can be used to collect the reflection power intensity under different weather conditions on the road at the same reflection power, calculate the reflection power intensity decay rate, and count the reflection power range of the detected vehicle and the reflection power intensity range of the water mist under different weather conditions. Using the reflection power intensity decay rate, the intersection of the reflection power intensity range of the laser radar scanning to the vehicle and the reflection power intensity range of the water mist is optimized when the laser radar is at a certain transmission power, reducing the influence of water mist on the vehicle when the vehicle passes through the detection range of the laser radar in rainy and snowy weather.

[0107] For target point cloud data, the first reflection intensity range corresponding to the first transmission power and the second reflection intensity range corresponding to the first transmission power can be adjusted according to the reflection intensity of the target point in the target vehicle point cloud data and the reflection intensity of other target points except the target vehicle point cloud data in the target point cloud data. The adjustment method of the reflection intensity range can include at least one of the following:

[0108] According to the reflection intensity of the target point in the target vehicle point cloud data, the reflection intensity range corresponding to the scanning point matched with the target point in the first reflection intensity range corresponding to the first transmission power is adjusted, the adjusted reflection intensity range contains the reflection intensity of the target point, and / or the reflection intensity range corresponding to the scanning point matched with the target point in the second reflection intensity range corresponding to the first transmission power is adjusted, the adjusted reflection intensity range does not contain the reflection intensity of the target point, or the adjusted reflection intensity range contains the reflection intensity of the target point, and the distance between the reflection intensity of the target point and the range boundary of the adjusted reflection intensity range is closer.

[0109] According to the reflection intensity of the target point in the target vehicle point cloud data, the reflection intensity range corresponding to the scanning point matched with the target point in the first reflection intensity range corresponding to the first transmission power is adjusted, the adjusted reflection intensity range contains the reflection intensity of the target point, and / or the reflection intensity range corresponding to the scanning point matched with the target point in the first reflection intensity range corresponding to the first transmission power is adjusted, the adjusted reflection intensity range does not contain the reflection intensity of the target point, or the adjusted reflection intensity range contains the reflection intensity of the target point, and the distance between the reflection intensity of the target point and the range boundary of the adjusted reflection intensity range is closer.

[0110] According to the first reflection intensity range corresponding to the first transmission power before adjustment and the first reflection intensity range corresponding to the first transmission power after adjustment, a reflection intensity correction value is determined, where the transmission intensity correction value is used to describe the change of the first reflection intensity range corresponding to the first transmission power, which can be a coefficient value, for example, 0.8, 1.2, etc.

[0111] For example, the water mist reflection intensity value corresponding to the water mist point cloud can be included in the reflection intensity range when the corresponding point is water mist, that is, the second reflection intensity range, and the reflection intensity range of the corresponding point in the first reflection intensity range is adjusted; according to the first reflection intensity range and the adjusted first reflection intensity range, the reflection intensity correction value is calculated, and the reflection intensity correction value is a coefficient value.

[0112] Using the reflection intensity correction value, the first reflection intensity range corresponding to other transmission powers in a group of transmission powers except the first transmission power can be adjusted to obtain the adjusted first reflection intensity range corresponding to the other transmission powers, and the reflection intensity range adjustment method can be the product of the first reflection intensity range and the reflection intensity correction value, that is, the first reflection intensity range is corrected to the following range: the product of the minimum reflection intensity of the first reflection intensity range and the reflection intensity correction value to the product of the maximum reflection intensity of the first reflection intensity range and the reflection intensity correction value.

[0113] Meanwhile, the second reflection intensity range corresponding to the other transmission power can be adjusted according to the first transmission power, the adjusted second reflection intensity range corresponding to the first transmission power, the other transmission power, and a reflection intensity attenuation rate, where the reflection intensity attenuation rate is used to describe the degree of attenuation of the signal intensity of the detection signal emitted by the laser radar in the process of transmission under different transmission powers.

[0114] For example, the reflection intensity range of the laser radar under other transmission powers in a rainy and snowy day can be corrected according to the reflection intensity correction value, and the reflection intensity range of the water mist under other powers can be calculated according to the water mist reflection intensity range and the reflection intensity attenuation rate.

[0115] According to the water mist reflection intensity range under the transmission power P i , the transmission power P i , the reflection intensity attenuation rate, and the transmission power P j , the reflection intensity range of the water mist under the transmission power P j is calculated, and the transmission power of the laser radar is adjusted to P j .

[0116] The water mist reflection intensity range under the transmission power P i is the fifth reflection intensity range, and the way to calculate the water mist reflection intensity range B under P j may be B=A*k(P i *(1-N i ) / (P j *(1-N j )), where A is the water mist reflection intensity range of P i , N i represents the attenuation rate under P i , k is a coefficient, and N j represents the attenuation rate under P j .

[0117] Through the embodiment, the first reflection intensity range and the second reflection intensity range corresponding to each transmission power are corrected based on the currently scanned point cloud data, which can improve the rationality of the transmission power adjustment.

[0118] In one example embodiment, the first transmission power is one of a set of transmission powers, and the set of transmission powers is similar to the foregoing embodiments, which will not be repeated here. Correspondingly, the above method further comprises:

[0119] S71, determine a light ray influence range corresponding to each of the set of transmission powers and a reflection intensity decay rate corresponding to each of the set of transmission powers, wherein the light ray influence range corresponding to each of the set of transmission powers is an intensity influence range of the laser radar on the intensity of the reflection intensity under each of the set of transmission powers;

[0120] S72, determine a fourth reflection intensity range corresponding to each of the set of transmission powers according to the third reflection intensity range corresponding to each of the set of transmission powers, the light ray influence range corresponding to each of the set of transmission powers, and the reflection intensity decay rate corresponding to each of the set of transmission powers, wherein the third reflection intensity range corresponding to each of the set of transmission powers is a reflection intensity range corresponding to each of the set of transmission powers when each of the scan points in the scan surface is a vehicle point under a non-preset weather state, the fourth reflection intensity range corresponding to each of the set of transmission powers is a reflection intensity range corresponding to each of the set of transmission powers when each of the scan points is a vehicle point under a preset weather state, and the target reflection intensity range corresponding to each of the target points is a reflection intensity range of the scan point corresponding to each of the target points in the fourth reflection intensity range corresponding to the first transmission power.

[0121] In the embodiment, the target reflection intensity range corresponding to each of the target points is a reflection intensity range corresponding to the first transmission power when the scan point corresponding to each of the target points is a vehicle point under a preset weather state, and the preset weather state is similar to that in the foregoing embodiment, which is not described herein. The reflection intensity range corresponding to each of the set of transmission powers when each of the scan points is a vehicle point under a preset weather state can be preset, so that the target reflection intensity range corresponding to each of the target points can be determined according to the following information: the reflection intensity range corresponding to each of the set of transmission powers when each of the scan points in the scan surface is a vehicle point under a non-preset weather state, the light ray influence range corresponding to each of the set of transmission powers (i.e., the intensity influence range of the laser radar on the intensity of the reflection intensity under each of the set of transmission powers), and the reflection intensity decay rate corresponding to each of the set of transmission powers. Here, the reflection intensity range corresponding to each of the set of transmission powers when each of the scan points is a vehicle point under a preset weather state can be the fourth reflection intensity range corresponding to each of the set of transmission powers.

[0122] In order to determine the fourth reflection intensity range corresponding to each of the set of transmission powers, the processing device (which can be a server or other processing device) can first determine the third reflection intensity range corresponding to each of the set of transmission powers, the light ray influence range corresponding to each of the set of transmission powers, and the reflection intensity decay rate corresponding to each of the set of transmission powers.

[0123] The third reflection intensity range corresponding to each transmission power can be obtained by scanning the vehicle by the laser radar, for example, the third reflection intensity range of the vehicle V1 scanned by the laser radar under each transmission power can be counted. il Here, l represents the vehicle category, and i represents the transmission power.

[0124] The light influence range corresponding to each transmission power of the laser radar can be configured based on experience or obtained by scanning by the laser radar, for example, the light influence range corresponding to each transmission power of the laser radar can be determined by comparing the reflection intensity of the ground point scanned by the laser radar under the preset weather and the non-preset weather.

[0125] The reflection intensity decay rate corresponding to each transmission power of the laser radar can be configured based on experience or obtained by scanning by the laser radar, for example, the reflection intensity decay rate corresponding to each transmission power of the laser radar can be fitted based on the reflection intensity of the ground point scanned by the laser radar under the preset weather and the non-preset weather, and the light influence range corresponding to each transmission power of the laser radar.

[0126] According to the obtained third reflection intensity range corresponding to each transmission power, the light influence range corresponding to each transmission power, and the reflection intensity decay rate corresponding to each transmission power, the fourth reflection intensity range corresponding to each transmission power can be calculated. For the case of distinguishing different vehicle categories, and based on the third reflection intensity range corresponding to each transmission power, the light influence range corresponding to each transmission power, and the reflection intensity decay rate corresponding to each transmission power under each vehicle category, the fourth reflection intensity range corresponding to each transmission power under each vehicle category can be calculated.

[0127] For example, the transmission intensity C j of the coarse classification vehicle V1 scanned by the laser radar under the non-rain and snow weather when the laser radar works at the transmission power P1 and at different distances L k and different angles β iljk in the road detection range can be established as a library, here, the scanning points in the scanning surface of the laser radar can be obtained by the distance L j and the angle β k ; the third reflection intensity range C il scanned by the laser radar on the different category vehicles V1 is counted, and the reflection intensity decay rate ηi and light influence range Δ i , the fourth reflection intensity range D of the laser radar on the vehicle V1 of different categories in the rain and snow weather is calculated il .

[0128] Through the embodiment, the reflection intensity range in the preset weather state is calculated according to the reflection intensity range in the non-preset weather state, the light influence range and the reflection intensity decay rate, and the convenience of determining the reflection intensity range can be improved.

[0129] In an example embodiment, the reflection intensity decay rate corresponding to each emission power of the laser radar is determined, comprising:

[0130] S81, in the case that no vehicle passes through the scanning surface and the weather state is a non-preset weather state, a first road surface reflection intensity range corresponding to each emission power of the laser radar is obtained, wherein the first road surface reflection intensity range corresponding to each emission power includes the reflection intensity range of each road surface point in the scanning surface matched with each emission power;

[0131] S82, in the case that no vehicle passes through the scanning surface and the weather state is a preset weather state, a second road surface reflection intensity range corresponding to each emission power of the laser radar is obtained, wherein the second road surface reflection intensity range corresponding to each emission power includes the reflection intensity range of each road surface point in the scanning surface matched with each emission power;

[0132] S83, the reflection intensity decay rate corresponding to each emission power is fitted according to the first road surface reflection intensity range corresponding to each emission power, the second road surface reflection intensity range corresponding to each emission power and the light influence range corresponding to each emission power.

[0133] In order to determine the reflection intensity decay rate corresponding to each emission power of the laser radar, the first road surface reflection intensity range corresponding to each emission power of the laser radar can be obtained in the case that no vehicle passes through the scanning surface and the weather state is a non-preset weather state. The first road surface reflection intensity range corresponding to each emission power here can include the reflection intensity range of each road surface point in the scanning surface matched with each emission power.

[0134] For example, the laser radar can select the emission power P i works, receives the reflection intensity of the road surface in the scanning detection range, which is S j , and the angle is α j .ij , the laser radar is counted to emit power P i When working, the first reflection intensity range A i (i.e., the first road surface reflection intensity range) of the road surface in the detection range is received.

[0135] The processing device can also obtain a second road surface reflection intensity range corresponding to each emission power, which is obtained by the laser radar scanning the scanning surface at each emission power when no vehicle passes through the scanning surface and the weather state is a preset weather state. The second road surface reflection intensity range corresponding to each emission power here can include a reflection intensity range of each road surface point in the scanning surface that matches each emission power.

[0136] For example, in rainy and snowy weather, the laser radar works at the same emission power P1, and the reflection intensity of the same distance and the same angle on the road surface in the detection range is B ij , the second reflection intensity range B i of the road surface at this emission power is counted. (i.e., the second road surface reflection intensity range). In addition, the range of the influence of external light on the reflection intensity when the laser radar scans the same distance and angle in the daytime and at night at this emission power can also be counted, so as to obtain a light influence range corresponding to each emission power.

[0137] The reflection intensity decay rate corresponding to each emission power can be fitted according to the first road surface reflection intensity range corresponding to each emission power, the second road surface reflection intensity range corresponding to each emission power, and the light influence range corresponding to each emission power. For example, by the first reflection intensity range, the second reflection intensity range, and the light influence range, the η n corresponding to the same emission power [P1, P2,..., P i is fitted, so that the reflection intensity decay rate [η1, η2,..., η n fitted at different emission powers of the laser radar is obtained.

[0138] Through the embodiment, the reflection intensity decay rate is fitted based on the road surface reflection intensity range in the non-preset weather state corresponding to each emission power, the road surface reflection intensity range in the preset weather state, and the light influence range, which can improve the convenience of determining the reflection intensity decay rate.

[0139] The vehicle information detection method in the embodiments of the application will be explained and described below in combination with optional examples. In the optional example, the abnormal weather is rainy and snowy weather, and the non-vehicle point is a water mist point.

[0140] To solve the problem that the laser radar is easy to detect water mist as a vehicle or part of a vehicle when detecting a vehicle in rainy and snowy weather, resulting in inaccurate vehicle data such as length, width, height, and even multiple vehicles or missing vehicles, the present alternative example provides a scheme for vehicle detection by a laser radar in rainy and snowy weather. The scheme establishes the reflection intensity range of the laser radar scanning a vehicle and the reflection intensity range of the laser radar scanning water mist, dynamically adjusts the emission power of the laser, and reduces the interference of water mist. Moreover, the detected point cloud data corresponds to the converted point cloud image, and a deep learning method is used to identify vehicles, water mist, etc. from the point cloud image. From the point cloud data corresponding to the point cloud image, water mist point cloud data is removed, water mist reflection intensity corresponding to the water mist point cloud data is collected and counted, vehicle point cloud data is extracted, and finally vehicle information is calculated. Through the above-mentioned manner, the cost of multi-sensor fusion is reduced, and the interference of water mist on the laser radar detecting a vehicle is also reduced, thereby improving the ability of the laser radar to detect a vehicle in rainy and snowy weather.

[0141] In combination Figure 3 As shown in the figure, the flow of the vehicle information detection method in the present alternative example can include the following steps:

[0142] Step 302, calculate and configure the reflection intensity range of the laser radar scanning different categories of vehicles under each emission power, which can be used as the reference reflection intensity range of each emission power scanning each category of vehicle. In addition, the reference reflection intensity range of each emission power scanning water mist can also be configured.

[0143] Step 304, when it is determined that the current weather is rainy and snowy, the expected vehicle (there may be a vehicle or there may be no vehicle) can be roughly classified based on the scanned point cloud data, and the expected vehicle category (roughly classified category) is obtained.

[0144] Step 306, based on the reflection intensity of each point in the detected point cloud data and the reference reflection intensity range of the expected vehicle category scanned under the current emission power, the water mist point cloud data in the scanned point cloud data is removed, and the remaining vehicle point cloud data is obtained.

[0145] In addition, point cloud image detection or point density distribution method can also be used to detect water mist and remove water mist.

[0146] Step 308, if the vehicle point cloud data is not empty, the vehicle information can be determined based on the vehicle point cloud data, otherwise, the process is ended.

[0147] In addition, the reference reflection intensity range of each type of vehicle scanned at each transmission power and the reference reflection intensity range of water mist scanned at each transmission power can also be updated using the scanned point cloud data, and the transmission power of the laser radar is dynamically adjusted based on the updated data. The transmission power of the laser radar can also be adjusted according to the amount of rain detected by the rain detector, that is, different rain ranges can correspond to different transmission powers, and the transmission power matching the amount of rain detected by the rain detector is determined as the adjusted transmission power.

[0148] Through the optional example, by dynamically adjusting the transmission power of the laser radar, the influence of rain and snow weather on the detection of vehicles by the laser radar can be reduced; at the same time, by dynamically adjusting the transmission power of the laser radar, performing water mist detection of the laser radar, and performing water mist elimination, the rain and snow weather can be determined only by using the laser radar, and the accuracy of the laser radar in detecting vehicles in rain and snow weather can be improved, and the anti-interference ability of the laser radar in rain and snow weather can be improved.

[0149] It should be noted that, for the foregoing method embodiments, in order to simply describe, they are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited to the action sequence described, because according to the present application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should know that the embodiments described in the specification all belong to preferred embodiments, and the actions and modules involved are not necessarily essential to the present application.

[0150] From the above description of the embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be realized by means of software and the necessary general hardware platform, and of course it can also be realized by hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium (such as a ROM (Read-Only Memory), a RAM (Random Access Memory), a magnetic disk, or an optical disk), and includes a number of instructions for causing an end device (which can be a mobile phone, a computer, a server, or a network device) to execute the method of each embodiment of the present application.

[0151] According to another aspect of the embodiments of the present application, a vehicle information detection system for implementing the vehicle information detection method described above is also provided, which can include:

[0152] The laser radar is used for scanning a scanning surface of the laser radar according to a first emission power to obtain target point cloud data and a reflection intensity of each target point in the target point cloud data in the case of an abnormal weather state.

[0153] The data processing component is configured to: obtain the target point cloud data and the reflection intensity of each target point; select target points with the reflection intensity in a corresponding target reflection intensity range from the target point cloud data and / or remove non-vehicle point cloud data from the target point cloud data based on a boundary frame to obtain target vehicle point cloud data, wherein the target reflection intensity range corresponding to each target point is a reflection intensity range matching the first emission power when the vehicle is scanned, and the boundary frame is a last data frame containing vehicle point cloud or a first data frame containing only non-vehicle point cloud after the last data frame containing vehicle point cloud in the target vehicle point cloud data determined according to the reflection intensity of each target point; and in the case that the target vehicle point cloud data is not empty, determine vehicle information of a target vehicle matching the target vehicle point cloud data according to the target vehicle point cloud data.

[0154] It should be noted that the data processing component in this embodiment can be used to execute the above steps S202, S204 and S206.

[0155] Through the above system, in the case of an abnormal weather state, the target point cloud data obtained by the laser radar scanning the scanning surface of the laser radar according to the first emission power and the reflection intensity of each target point in the target point cloud data are obtained; target points with the reflection intensity in a corresponding target reflection intensity range are selected from the target point cloud data and / or non-vehicle point cloud data is removed from the target point cloud data based on a boundary frame to obtain target vehicle point cloud data, wherein the target reflection intensity range corresponding to each target point is a reflection intensity range matching the first emission power when the vehicle is scanned, and the boundary frame is a last data frame containing vehicle point cloud or a first data frame containing only non-vehicle point cloud after the last data frame containing vehicle point cloud in the target vehicle point cloud data determined according to the reflection intensity of each target point; and in the case that the target vehicle point cloud data is not empty, vehicle information of a target vehicle matching the target vehicle point cloud data is determined according to the target vehicle point cloud data, which solves the technical problem in the related art that the detection method of vehicle information has low accuracy due to the influence of interference particles in the air, and improves the accuracy of vehicle information detection.

[0156] In an example embodiment, the data processing component is further configured to determine a reference intensity value of each scan point in the scan plane, wherein the reference intensity value of each scan point is an average value of a range of reflected intensity matching the first transmission power when each scan point is the vehicle point; and perform the boundary frame determination step cyclically until a loop end condition is satisfied, wherein the loop end condition comprises that N is 2 or a boundary frame is determined, and the boundary frame determination step comprises: sequentially selecting N continuous frames from the multi-frame point cloud data included in the target point cloud data, wherein N is a positive integer greater than 2; determining missing points of each frame in the N continuous frames, wherein the missing points of each frame are target points included in at least one frame other than each frame in the N continuous frames and not included in each frame; expanding the missing points of each frame using the reference intensity value of each scan point to obtain an expanded each frame, wherein the points included in the expanded each frame are the same; determining a reflected intensity average value frame, wherein the points included in the reflected intensity average value frame are the same as the points included in the expanded each frame, and the reflected intensity of each point in the reflected intensity average value frame is an average value of the reflected intensity of the corresponding point in the expanded each frame; determining a variance of the reflected intensity of the points in each frame before expansion and the reflected intensity of the corresponding point in the reflected intensity average value frame to obtain a reflected intensity variance corresponding to each frame; in a case where there is a first frame in the N continuous frames corresponding to the reflected intensity variance greater than or equal to a first variance threshold, the reflected intensity variances corresponding to the first frame to the last frame in the multi-frame point cloud data are all greater than or equal to the first variance threshold, and the variance of the corresponding points of the first frame and the frame before the first frame is greater than or equal to a second variance threshold, the first frame is determined as the boundary frame; and in a case where no boundary frame is determined from the multi-frame point cloud data, updating N using a value obtained after N is reduced by 1 to obtain an updated N.

[0157] In an example embodiment, the abnormal weather state is a rainy and snowy weather state; and the data processing component is further configured to determine that the current is in the rainy and snowy weather state in a case where the point cloud data of K continuous vehicles all include water mist point cloud data, wherein K is a positive integer greater than or equal to 2; and determine that the current is in a non-rainy and snowy weather state in a case where the point cloud data of M continuous vehicles all do not include water mist point cloud data or the variance of the reflected intensity of each road surface point in the scan plane in T continuous frames is greater than or equal to a third variance threshold, wherein M and T are both positive integers greater than or equal to 2.

[0158] In an example embodiment, the data processing component is further configured to determine a vehicle start frame and a vehicle end frame corresponding to the detected current vehicle; determine a vehicle profile parameter corresponding to each frame in the vehicle start frame to the vehicle end frame, wherein the vehicle profile parameter comprises at least one of a vehicle height, a vehicle width; determine that the point cloud data of the current vehicle contains the fog point cloud data in a case that the vehicle profile parameter corresponding to each of the Q consecutive frames is less than or equal to a preset parameter threshold, wherein Q is a positive integer greater than or equal to 2; or determine a reflection intensity variance corresponding to each frame in the vehicle start frame to the vehicle end frame, wherein the reflection intensity variance corresponding to each frame is a variance of reflection intensity of points in each frame and corresponding points in a previous frame of each frame; determine that the point cloud data of the current vehicle contains the fog point cloud data in a case that there is a second frame corresponding to a reflection intensity variance greater than or equal to a fourth variance threshold in the vehicle start frame to the vehicle end frame, the reflection intensity variances corresponding to L consecutive frames starting from the second frame are less than or equal to a fifth variance threshold, and the variance of reflection intensity of points in the L consecutive frames and corresponding points in a previous frame of the second frame is greater than or equal to the fourth variance threshold, wherein L is a positive integer greater than or equal to 2.

[0159] In an example embodiment, the first transmission power is one of a set of transmission powers; and the data processing component is further configured to select a second transmission power from the set of transmission powers according to a first reflection intensity range corresponding to each transmission power in the set of transmission powers and a second reflection intensity range corresponding to each transmission power, wherein the first reflection intensity range corresponding to each transmission power is a reflection intensity range matched to each transmission power when scanning the vehicle, and the second reflection intensity range corresponding to each transmission power is a reflection intensity range matched to each transmission power when scanning the non-vehicle point; and adjust the transmission power of the lidar from the first transmission power to the second transmission power.

[0160] In an example embodiment, the data processing component is further configured to select a second transmission power from the set of transmission powers according to a range intersection between the first reflection intensity range corresponding to each transmission power and the second reflection intensity range corresponding to each transmission power, and a range size of the first reflection intensity range corresponding to each transmission power.

[0161] In an example embodiment, the data processing component is further configured to, before selecting the second transmit power from the set of transmit powers according to the first reflection intensity range corresponding to the first transmit power and the second reflection intensity range corresponding to the first transmit power, adjust the first reflection intensity range corresponding to the first transmit power and the second reflection intensity range corresponding to the first transmit power according to reflection intensities of target points in the target vehicle point cloud data and reflection intensities of other target points in the target point cloud data other than the target vehicle point cloud data; determine a reflection intensity correction value according to the first reflection intensity range corresponding to the first transmit power before adjustment and the first reflection intensity range corresponding to the first transmit power after adjustment; adjust the first reflection intensity ranges corresponding to the other transmit powers in the set of transmit powers other than the first transmit power using the reflection intensity correction value to obtain the first reflection intensity ranges corresponding to the other transmit powers after adjustment; and adjust the second reflection intensity ranges corresponding to the other transmit powers according to the first transmit power, the second reflection intensity range corresponding to the first transmit power after adjustment, the other transmit powers, and the reflection intensity decay rate to obtain the second reflection intensity ranges corresponding to the other transmit powers after adjustment.

[0162] In an example embodiment, the first transmit power is one of the set of transmit powers; and the data processing component is further configured to determine a light ray influence range of the lidar corresponding to each transmit power in the set of transmit powers, wherein the light ray influence range corresponding to each transmit power is an intensity influence range of the lidar on reflection intensity of light rays under each transmit power; determine a reflection intensity decay rate of the lidar corresponding to each transmit power; and determine a fourth reflection intensity range corresponding to each transmit power according to a third reflection intensity range corresponding to each transmit power, the light ray influence range corresponding to each transmit power, and the reflection intensity decay rate corresponding to each transmit power, wherein the third reflection intensity range corresponding to each transmit power is a reflection intensity range corresponding to each transmit power when each scan point in the scan surface is a vehicle point under a non-exceptional weather state, the fourth reflection intensity range corresponding to each transmit power is a reflection intensity range corresponding to each transmit power when each scan point is a vehicle point under an exceptional weather state, and the target reflection intensity range corresponding to each target point is a reflection intensity range of the scan point corresponding to each target point in the fourth reflection intensity range corresponding to the first transmit power.

[0163] In an example embodiment, the data processing component is further configured to, in a case that no vehicle passes the scanning surface and the weather state is a non-exceptional weather state, acquire a first road surface reflection intensity range corresponding to each of the transmission powers, which is obtained by the laser radar scanning the scanning surface according to each of the transmission powers, wherein the first road surface reflection intensity range corresponding to each of the transmission powers comprises a reflection intensity range of each road surface point in the scanning surface matching each of the transmission powers; in a case that no vehicle passes the scanning surface and the weather state is an exceptional weather state, acquire a second road surface reflection intensity range corresponding to each of the transmission powers, which is obtained by the laser radar scanning the scanning surface according to each of the transmission powers, wherein the second road surface reflection intensity range corresponding to each of the transmission powers comprises a reflection intensity range of each road surface point in the scanning surface matching each of the transmission powers; and perform reflection intensity decay rate fitting according to the first road surface reflection intensity range corresponding to each of the transmission powers, the second road surface reflection intensity range corresponding to each of the transmission powers, and a light ray influence range corresponding to each of the transmission powers, to obtain a reflection intensity decay rate corresponding to each of the transmission powers.

[0164] In an example embodiment, the laser radar is a transverse scanning laser radar.

[0165] It should be noted that the above modules and the examples and application scenarios implemented by the corresponding steps are the same as those disclosed in the above embodiments, but are not limited to the above disclosed content. It should be noted that the above modules as part of the device can run in the hardware environment as shown in Figure 1 The hardware environment includes a network environment.

[0166] According to another aspect of the embodiments of the present application, a storage medium is provided. Optionally, in the present embodiment, the storage medium can be used to execute the program code of any of the above vehicle information detection methods.

[0167] Optionally, in the present embodiment, the storage medium can be located on at least one of the plurality of network devices in the network shown in the above embodiments.

[0168] Optionally, in the present embodiment, the storage medium is configured to store program code for executing the following steps:

[0169] S1, in the case of an exceptional weather state, acquiring target point cloud data obtained by the laser radar scanning the scanning surface of the laser radar according to a first transmission power and the reflection intensity of each target point in the target point cloud data;

[0170] S2, select target points with reflection intensity in a corresponding target reflection intensity range from the target point cloud data and / or remove non-vehicle point cloud data from the target point cloud data based on a boundary frame to obtain target vehicle point cloud data, wherein the target reflection intensity range corresponding to each target point is a reflection intensity range matching the first transmission power when the vehicle is scanned, and the boundary frame is a last data frame containing vehicle point cloud or a first data frame containing only non-vehicle point cloud after the last data frame containing vehicle point cloud in the target vehicle point cloud data determined according to the reflection intensity of each target point;

[0171] S3, in the case that the target vehicle point cloud data is not empty, determining vehicle information of a target vehicle matching the target vehicle point cloud data according to the target vehicle point cloud data.

[0172] Optionally, specific examples in the embodiment can refer to examples described in the above embodiments, and details are not described herein.

[0173] Optionally, in the embodiment, the storage medium can include but is not limited to a U disk, a ROM, a RAM, a mobile hard disk, a magnetic disk or an optical disk and various storage program codes.

[0174] According to another aspect of the embodiment of the application, an electronic device for implementing the above vehicle information detection method is also provided, which can be a server, a terminal or a combination thereof.

[0175] Figure 4 is a structural block diagram of an optional electronic device according to the embodiment of the application, as shown in Figure 4 including a processor 402, a communication interface 404, a memory 406 and a communication bus 408, wherein the processor 402, the communication interface 404 and the memory 406 complete mutual communication through the communication bus 408, wherein,

[0176] the memory 406 is configured to store a computer program;

[0177] the processor 402 is configured to execute the computer program stored in the memory 406 to implement the following steps:

[0178] S1, in the case of abnormal weather state, obtaining target point cloud data obtained by scanning a scanning surface of a laser radar according to a first transmission power of the laser radar and reflection intensity of each target point in the target point cloud data;

[0179] S2, selecting a target point with a reflection intensity in a corresponding target reflection intensity range from the target point cloud data and / or removing non-vehicle point cloud data from the target point cloud data based on a boundary frame to obtain target vehicle point cloud data, wherein the target reflection intensity range corresponding to each target point is a reflection intensity range matched with the first transmission power when the vehicle is scanned, and the boundary frame is a last data frame containing vehicle point cloud or a first data frame containing only non-vehicle point cloud after the last data frame containing vehicle point cloud in the target vehicle point cloud data determined according to the reflection intensity of each target point.

[0180] S3, in the case that the target vehicle point cloud data is not empty, determining vehicle information of a target vehicle matched with the target vehicle point cloud data according to the target vehicle point cloud data.

[0181] Optionally, the communication bus can be a PCI (Peripheral Component Interconnect) bus, an EISA (Extended Industry Standard Architecture) bus, or the like. The communication bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, Figure 4 Only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus. The communication interface is used for communication between the electronic device and other devices.

[0182] The memory can include a RAM and can also include a non-volatile memory, for example, at least one disk memory. Optionally, the memory can also be at least one storage device located away from the aforementioned processor.

[0183] The aforementioned processor can be a general-purpose processor, which can include but is not limited to a CPU (Central Processing Unit), an NP (Network Processor), and the like; and can also be a DSP (Digital Signal Processing), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component.

[0184] Optionally, specific examples in the present embodiment can refer to examples described in the above-described embodiments, and the present embodiment will not be described here again.

[0185] Those skilled in the art can understand that, Figure 4 The structure shown is only schematic, and the device implementing the method for detecting vehicle information can be a terminal device, which can be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a palm computer, a Mobile Internet Device (MID), a PAD, or the like. Figure 4 This does not limit the structure of the electronic device. For example, the electronic device can further include more or fewer components (such as a network interface, a display device, etc.) than Figure 4 shown in the description, or have a different configuration from Figure 4 shown.

[0186] Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device by a program, which can be stored in a computer readable storage medium, which can include a flash disk, a ROM, a RAM, a magnetic disk or an optical disk, etc.

[0187] The serial numbers of the embodiments of the present application described above are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0188] The integrated units in the above embodiments, if realized in the form of software function units and sold or used as independent products, can be stored in the above computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the parts that make contributions to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing one or more computer devices (which can be personal computers, servers or network devices, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application.

[0189] In the above embodiments of the present application, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0190] In the several embodiments provided by the present application, it should be understood that the disclosed client can be implemented in other ways. Of course, the device embodiment described above is only schematic. For example, the division of the units is only a logical function division, and there can be another division manner in actual implementation. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interface, unit or module, and can be electrical or other forms.

[0191] The units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, may be located in one place or distributed on multiple network units. Part or all of the units can be selected to achieve the purpose of the scheme provided in the embodiment according to actual needs.

[0192] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically alone, or at least two units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0193] The above is only the preferred embodiment of the present application, and it should be pointed out that for ordinary skilled persons in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, and these improvements and refinements should be considered as the protection scope of the present application.

Claims

1. A method for detecting vehicle information, characterized in that: include: In an abnormal weather condition, obtaining target point cloud data obtained by scanning a scanning surface of the laser radar at a first transmission power and a reflection intensity of each target point in the target point cloud data, wherein the abnormal weather condition is rain or snow; Selecting target points whose reflection intensities are within a corresponding target reflection intensity range from the target point cloud data and / or eliminating non-vehicle point cloud data from the target point cloud data based on a boundary frame, to obtain target vehicle point cloud data, wherein the target reflection intensity range corresponding to each target point is a reflection intensity range that matches the first transmission power when a vehicle is scanned, and the boundary frame is determined based on the reflection intensity of each target point, and is the last data frame containing a vehicle point cloud or the first data frame containing only non-vehicle point clouds after the last data frame containing a vehicle point cloud in the target vehicle point cloud data; In a case where the target vehicle point cloud data is not empty, determining vehicle information of the target vehicle that matches the target vehicle point cloud data according to the target vehicle point cloud data; When it is determined that water mist point cloud data exists in the point cloud data of K consecutive vehicles, it is determined that the current weather condition is rainy or snowy, where K is a positive integer greater than or equal to 2; when it is determined that water mist point cloud data does not exist in the point cloud data of M consecutive vehicles, or the variance of the reflection intensity of each road surface point in the scanning surface in T consecutive frames is greater than or equal to a third difference threshold, it is determined that the current weather condition is not rainy or snowy, where M and T are both positive integers greater than or equal to 2.

2. The method according to claim 1, characterized in that The method further comprises: Determining a reference intensity value for each scanning point within the scanning surface, wherein the reference intensity value for each scanning point is an average value of a reflection intensity range that matches the first transmission power when each scanning point is a vehicle point; The boundary frame determination step is executed cyclically until a loop end condition is satisfied, wherein the loop end condition includes N being 2 or the boundary frame being determined, and the boundary frame determination step includes: Sequentially selecting N consecutive frames from the multiple frames of point cloud data included in the target point cloud data, where N is a positive integer greater than 2; Determining a missing point in each of the N consecutive frames, wherein the missing point in each frame is a target point included in at least one frame other than each of the N consecutive frames and not included in each of the frames; Using the reference intensity value of each scanning point to expand the missing points of each frame to obtain each expanded frame, wherein each expanded frame includes the same points; Determining a reflection intensity mean frame, wherein the points included in the reflection intensity mean frame are the same as the points included in each of the expanded frames, and the reflection intensity of each point in the reflection intensity mean frame is the average of the reflection intensities of the corresponding points in each of the expanded frames; Determine the variance of the reflection intensity of a point in each frame before expansion and the reflection intensity of a corresponding point in the reflection intensity mean frame to obtain the reflection intensity variance corresponding to each frame; If there is a first frame among the N consecutive frames whose corresponding reflection intensity variance is greater than or equal to a first variance threshold, the reflection intensity variances corresponding to the first frame to the last frame in the multi-frame point cloud data are all greater than or equal to the first variance threshold, and the variances of corresponding points of the first frame and the frame before the first frame are greater than or equal to a second variance threshold, the first frame is determined as the boundary frame; In the case that the boundary frame is not determined from the multi-frame point cloud data, N is updated using a value obtained by subtracting 1 from N to obtain an updated N.

3. The method according to claim 1, characterized in that The method further comprises: Determining a vehicle start frame and a vehicle end frame corresponding to the detected current vehicle; Determine a vehicle contour parameter corresponding to each frame from the vehicle start frame to the vehicle end frame, wherein the vehicle contour parameter includes at least one of the following: vehicle height, vehicle width; if there are Q consecutive frames whose corresponding vehicle contour parameters are all less than or equal to a preset parameter threshold, determine that water mist point cloud data exists in the point cloud data of the current vehicle, wherein Q is a positive integer greater than or equal to 2; or Determine the reflection intensity change variance corresponding to each frame from the vehicle start frame to the vehicle end frame, wherein the reflection intensity change variance corresponding to each frame is the variance of the reflection intensity between the point in each frame and the corresponding point in the previous frame of each frame; when there is a second frame from the vehicle start frame to the vehicle end frame whose corresponding reflection intensity change variance is greater than or equal to a fourth variance threshold, the reflection intensity change variance corresponding to L consecutive frames starting from the second frame is less than or equal to a fifth variance threshold, and the variance of the reflection intensity between the point in the consecutive L frames and the corresponding point in the previous frame of the second frame is greater than or equal to the fourth variance threshold, determine that water mist point cloud data exists in the point cloud data of the current vehicle, wherein L is a positive integer greater than or equal to 2.

4. The method according to claim 1, wherein The first transmit power is one of a group of transmit powers; and the method further includes: Selecting a second transmission power from the set of transmission powers based on a first reflection intensity range corresponding to each transmission power in the set of transmission powers and a second reflection intensity range corresponding to each transmission power, wherein the first reflection intensity range corresponding to each transmission power is a reflection intensity range that matches each transmission power when scanning a vehicle, and the second reflection intensity range corresponding to each transmission power is a reflection intensity range that matches each transmission power when scanning a non-vehicle point; Adjust the transmit power of the laser radar from the first transmit power to the second transmit power.

5. The method according to claim 4, characterized in that The selecting the second transmission power from the set of transmission powers according to the first reflection intensity range corresponding to each transmission power and the second reflection intensity range corresponding to each transmission power includes: The second transmission power is selected from the group of transmission powers based on the range intersection between the first reflection intensity range corresponding to each transmission power and the second reflection intensity range corresponding to each transmission power, and the range size of the first reflection intensity range corresponding to each transmission power.

6. The method according to claim 4, characterized in that Before selecting the second transmission power from the set of transmission powers based on the first reflection intensity range corresponding to each transmission power and the second reflection intensity range corresponding to each transmission power, the method further includes: Adjusting a first reflection intensity range corresponding to the first transmission power and a second reflection intensity range corresponding to the first transmission power according to the reflection intensity of the target point in the target vehicle point cloud data and the reflection intensity of other target points in the target point cloud data except the target vehicle point cloud data; determining a reflection intensity correction value according to a first reflection intensity range corresponding to the first transmission power before adjustment and the first reflection intensity range corresponding to the first transmission power after adjustment; Using the reflection intensity correction value, adjust the first reflection intensity ranges corresponding to the other transmission powers in the set of transmission powers except the first transmission power to obtain adjusted first reflection intensity ranges corresponding to the other transmission powers; According to the first transmission power, the adjusted second reflection intensity range corresponding to the first transmission power, the other transmission powers and the reflection intensity attenuation rate, the second reflection intensity range corresponding to the other transmission powers is adjusted to obtain the adjusted second reflection intensity range corresponding to the other transmission powers.

7. The method according to claim 1, characterized in that The first transmit power is one of a group of transmit powers; and the method further includes: Determining a light influence range of the laser radar corresponding to each transmission power in the set of transmission powers, and determining a reflection intensity attenuation rate of the laser radar corresponding to each transmission power, wherein the light influence range corresponding to each transmission power is an intensity influence range of the interference light of the laser radar on the reflection intensity at each transmission power; Based on the third reflection intensity range corresponding to each transmission power, the light influence range corresponding to each transmission power, and the reflection intensity attenuation rate corresponding to each transmission power, the fourth reflection intensity range corresponding to each transmission power is determined, wherein the third reflection intensity range corresponding to each transmission power is the reflection intensity range that matches each transmission power under non-abnormal weather conditions when each scanning point in the scanning surface is a vehicle point, the fourth reflection intensity range corresponding to each transmission power is the reflection intensity range that matches each transmission power under abnormal weather conditions when each scanning point is a vehicle point, and the target reflection intensity range corresponding to each target point is the reflection intensity range of the scanning point corresponding to each target point in the fourth reflection intensity range corresponding to the first transmission power.

8. The method according to claim 7, characterized in that Determining the reflection intensity attenuation rate corresponding to each transmission power of the laser radar includes: When no vehicle passes through the scanning surface and the weather condition is the non-abnormal weather condition, obtaining a first road surface reflection intensity range corresponding to each transmission power, obtained by the laser radar scanning the scanning surface according to each transmission power, wherein the first road surface reflection intensity range corresponding to each transmission power includes the reflection intensity range of each road surface point within the scanning surface that matches each transmission power; When no vehicle passes through the scanning surface and the weather condition is the abnormal weather condition, obtaining a second road surface reflection intensity range corresponding to each transmission power, obtained by the laser radar scanning the scanning surface according to each transmission power, wherein the second road surface reflection intensity range corresponding to each transmission power includes a reflection intensity range of each road surface point within the scanning surface that matches each transmission power; The reflection intensity attenuation rate is fitted according to the first road surface reflection intensity range corresponding to each transmission power, the second road surface reflection intensity range corresponding to each transmission power, and the light influence range corresponding to each transmission power to obtain the reflection intensity attenuation rate corresponding to each transmission power.

9. A vehicle information detection system, characterized in that: include: a laser radar, configured to scan a scanning surface of the laser radar at a first transmission power in an abnormal weather condition to obtain target point cloud data and a reflection intensity of each target point in the target point cloud data, wherein the abnormal weather condition is rain or snow; a data processing component for acquiring the target point cloud data and the reflection intensity of each target point; selecting target points whose reflection intensity is within a corresponding target reflection intensity range from the target point cloud data and / or eliminating non-vehicle point cloud data from the target point cloud data based on a boundary frame, to obtain target vehicle point cloud data, wherein the target reflection intensity range corresponding to each target point is a reflection intensity range that matches the first transmission power when a vehicle is scanned, and the boundary frame is determined based on the reflection intensity of each target point, and is the last data frame containing a vehicle point cloud or the first data frame containing only non-vehicle point clouds after the last data frame containing a vehicle point cloud in the target vehicle point cloud data; and when the target vehicle point cloud data is not empty, determining vehicle information of the target vehicle that matches the target vehicle point cloud data based on the target vehicle point cloud data; Wherein, the data processing component is also used to determine that the current weather condition is rainy or snowy when it is determined that water mist point cloud data exists in the point cloud data of K consecutive vehicles, where K is a positive integer greater than or equal to 2; and to determine that the current weather condition is not rainy or snowy when it is determined that water mist point cloud data does not exist in the point cloud data of M consecutive vehicles, or the variance of the reflection intensity of each road surface point in the scanning surface in T consecutive frames is greater than or equal to a third difference threshold, where M and T are both positive integers greater than or equal to 2.

Images