Vehicle ranging system and vehicle

Abstract

The utility model provides a vehicle ranging system and a vehicle. The vehicle ranging system includes: a vehicle body; the at least two binocular cameras are arranged on a vehicle front cover of the vehicle body, and the binocular cameras are used for acquiring real-time images in the advancing direction of the vehicle body; the at least two single-point laser modules are arranged on a vehicle front cover of the vehicle body, the at least two single-point laser modules and the at least two binocular cameras are arranged in a one-to-one correspondence mode, the single-point laser modules are located between the binocular cameras and a front windshield of the vehicle body, and the single-point laser modules are used for irradiating objects corresponding to fixed pixel points of real-time images. The distance of the object is detected. According to the vehicle distance measurement system, the distance measurement effect can be effectively improved.

Description

Technical Field

[0001] This application relates to the field of driverless technologies, and particularly to a vehicle ranging system and a vehicle. Background Art

[0002] Driverless technology refers to a technical system that can automatically complete operations such as vehicle navigation, obstacle avoidance, acceleration, and braking without direct intervention by a human driver. In driverless driving, it is necessary to measure the distance to the object in front to control the driving of the vehicle according to the distance.

[0003] In the ranging methods in the related art, the ranging effect is poor. Utility Model Content

[0004] Embodiments of this application provide a vehicle ranging system and a vehicle.

[0005] In a first aspect, embodiments of this application provide a vehicle ranging system, including:

[0006] A vehicle body;

[0007] At least two binocular cameras, provided on the front cover of the vehicle body, and the binocular cameras are used to obtain real-time images in the traveling direction of the vehicle body;

[0008] At least two single-point laser modules, provided on the front cover of the vehicle body, at least two of the single-point laser modules are arranged in one-to-one correspondence with at least two of the binocular cameras, the single-point laser modules are located between the binocular cameras and the front windshield of the vehicle body, and the single-point laser modules are used to irradiate the object corresponding to the fixed pixel points of the real-time image to detect the distance of the object.

[0009] In an embodiment, the single-point laser module includes at least three rows of single-point lidars, and the detection distances of the single-point lidars in different rows are different.

[0010] In an embodiment, the two binocular cameras are symmetrically arranged about the midline of the vehicle body, and the two single-point laser modules are symmetrically arranged about the midline of the vehicle body.

[0011] In an embodiment, the single-point laser module includes a plurality of single-point lidars, and different single-point lidars are used to irradiate and align the objects corresponding to different pixel points of the real-time image.

[0012] In an embodiment, the binocular camera includes a binocular camera base and a binocular camera body, the binocular camera base is detachably connected to the front cover, and the binocular camera body is connected to the binocular camera base.

[0013] In one embodiment, along the direction from the binocular camera to the single-point laser module, the cross-sectional area of the binocular camera base gradually increases so that the binocular camera base is attached to the front hood of the vehicle.

[0014] In one embodiment, the single-point laser module includes a single-point laser base and a single-point laser body. The single-point laser base is detachably connected to the front hood of the vehicle, and the single-point laser body is connected to the single-point laser base.

[0015] In one embodiment, along the direction from the binocular camera to the single-point laser module, the cross-sectional area of the single-point laser base gradually increases so that the single-point laser base is attached to the front hood of the vehicle.

[0016] In one embodiment, the shooting area of one of the binocular cameras is adjacent to or partially overlaps with the shooting area of the other binocular camera.

[0017] In a second aspect, an embodiment of the present application provides a vehicle, including the vehicle ranging system as described above.

[0018] Advantages of the embodiments of the present application:

[0019] In the embodiments of the present application, a single-point laser module is correspondingly arranged at each binocular camera, and precise ranging of the object in front of the vehicle body is realized through the combination of the binocular camera and the single-point laser module. Moreover, by arranging the binocular camera and the single-point laser module on the front hood of the vehicle, the shooting angle is kept consistent with the driving angle of the vehicle, which can better detect the object in front. At the same time, by setting at least two binocular cameras and the corresponding single-point laser modules, the shooting interface is expanded, the detection range is expanded, and the ranging effect can be effectively improved. Description of the Drawings

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present application. For those skilled in the art, without creative efforts, other drawings can be obtained according to these drawings.

[0021] Figure 1 is a flowchart of the vehicle ranging method provided by the embodiment of the present application;

[0022] Figure 2 is a schematic structural diagram of the vehicle ranging device provided by the embodiment of the present application;

[0023] Figure 3 is a schematic structural diagram of the vehicle ranging system provided by the embodiment of the present application;

[0024] Figure 4 It is a partial structural schematic diagram of a vehicle ranging system provided by an embodiment of the present application. B1, B2, B3, and B4 in the figure represent different pixel points;

[0025] Figure 5 It is a structural schematic diagram of an electronic device provided by an embodiment of the present application. Specific embodiments

[0026] Next, the technical solutions in the embodiments of the present application will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without making creative efforts belong to the scope of protection of the present application. In addition, it should be understood that the specific embodiments described herein are only used to illustrate and explain the present application, and are not used to limit the present application. In the present application, unless otherwise stated, the orientation words such as "upper" and "lower" generally refer to the upper and lower in the actual use or working state of the device, specifically the drawing direction in the accompanying drawings; and "inner" and "outer" refer to the outline of the device.

[0027] Next, in combination with Figures 1 to 5 Describe the vehicle ranging system and vehicle of the present application.

[0028] According to an embodiment of the first aspect of the present application, as Figure 1 , the vehicle includes a binocular camera 2 and a plurality of single-point lidars 31. The single-point lidar 31 is adapted to emit ranging laser. The vehicle ranging method includes:

[0029] Step 101: Use the binocular camera 2 to obtain the light reflected by an object in the traveling direction of the vehicle to obtain a real-time image;

[0030] It can be understood that after the binocular camera 2 obtains the light reflected by the object, a multi-spectral image in the traveling direction can be obtained based on the light.

[0031] It can be understood that the traveling direction may include the front-back direction, and the front-back direction may exceed a range of 180 degrees.

[0032] In some examples, the binocular camera 2 includes a multi-channel multi-spectral chip. The multi-channel spectral chip is used to obtain lights of different wavelengths reflected by an object in the traveling direction to obtain a real-time multi-spectral image. Among them, the central wavelengths corresponding to different channels are different, so that different channels obtain spectral information of different wavelengths. Specifically, the multi-channel spectral chip can obtain natural light reflected by the object. It can also be that different wavelengths of light are first emitted to the object through a light source, and the multi-channel spectral chip obtains the light emitted by the light source reflected by the object.

[0033] Step 102: Obtain the emission and reception durations of multiple single-point lidars 31, where different single-point lidars 31 correspond to objects corresponding to different pixel points of the real-time image being illuminated.

[0034] It can be understood that each single-point lidar 31 is controlled to illuminate objects corresponding to different pixel points on the real-time image obtained by the binocular camera 2, that is, different single-point lidars 31 correspond to illuminating actual objects corresponding to different pixel points, and the emission and reception durations of each single-point lidar 31 are obtained.

[0035] Step 103: Determine the object illuminated by the target single-point lidar 31 as the calibration point, where the emission and reception duration of the target single-point lidar 31 is equal to the standard duration, and the emission and reception duration of the single-point lidar 31 when illuminating an object at the standard distance is the standard duration.

[0036] It can be understood that the emission and reception durations of each single-point lidar 31 are compared with the standard duration, and the object illuminated by the single-point lidar 31 whose emission and reception duration is equal to the standard duration is determined as the calibration point.

[0037] The single-point lidar 31 whose emission and reception duration is equal to the standard duration is described as the standard single-point lidar 31. The emission and reception duration of the standard single-point lidar 31 when illuminating the corresponding calibration point is equal to the standard duration, and the emission and reception duration of the single-point lidar 31 when illuminating an object at the standard distance is the standard duration. That is to say, the actual distance of the calibration point is the standard distance.

[0038] Step 104: Determine the binocular camera measurement distance of the object to be measured in the real-time image.

[0039] It can be understood that the binocular camera 2 is used to measure the distance of the object to be measured to obtain the binocular camera measurement distance.

[0040] Exemplarily, the binocular geometric algorithm can be used to determine the binocular camera measurement distance of the object to be measured.

[0041] Step 105: Calibrate the binocular camera measurement distance based on the image coordinates of the object to be measured, the image coordinates of the calibration point, and the standard distance to obtain the actual distance.

[0042] It can be understood that the actual distance of the calibration point is the standard distance, that is, the actual distance of the calibration point is known. At the same time, the image coordinates of the object to be measured and the calibration point, that is, the coordinates of the object to be measured and the calibration point in the real-time image, can also be calculated. Furthermore, based on the image coordinates of the object to be measured, the image coordinates of the calibration point, and the standard distance, the binocular camera measurement distance can be calibrated to obtain the actual distance of the object to be measured.

[0043] In some examples, after step 103, it is also possible to determine whether the central pixel point of the object to be measured is a calibration point. If the central pixel point of the object to be measured is the calibration point, it means that the actual distance of the object to be measured at this time is the standard distance, and thus steps 104 and 105 do not need to be performed.

[0044] According to the vehicle ranging method of the embodiment of the present application, real-time images in the vehicle traveling direction are obtained through the binocular camera 2; different single-point lidars 31 are controlled to irradiate objects corresponding to different pixel points of the real-time images obtained by the binocular camera 2, and the emission and reception durations of different single-point lidars 31 are obtained. Since the emission and reception duration of the single-point lidar 31 irradiating an object at the standard distance is the standard duration, the object corresponding to the single-point lidar 31 with the emission and reception duration equal to the standard duration is determined as the calibration point; the distance of the object to be measured is measured by the binocular camera 2 to obtain a binocular camera measurement distance, and then the binocular camera measurement distance is calibrated by using the image coordinates of the object to be measured, the image coordinates of the calibration point, and the distance of the calibration point (i.e., the standard distance) to obtain the actual distance, improving the ranging accuracy.

[0045] When calibrating the binocular camera 2 measurement distance based on the coordinates of the object to be measured, the coordinates of the calibration point, and the standard distance to obtain the actual distance, the distance between the object to be measured and the calibration point can be first calculated according to the image coordinates of the object to be measured and the image coordinates of the calibration point, and then combined with the actual distance of the calibration point (i.e., the standard distance), the actual distance of the object to be measured can be calculated, and thus the binocular camera measurement distance can be calibrated.

[0046] When calibrating the binocular camera measurement distance based on the coordinates of the object to be measured, the coordinates of the calibration point, and the standard distance to obtain the actual distance, it is also possible to obtain the binocular measurement distance of the calibration point by using the binocular camera 2, compare and calculate the actual distance of the calibration point (i.e., the standard distance) and the binocular measurement distance of the calibration point, and a calibration coefficient can be obtained, that is, the binocular measurement distance of the calibration point * calibration coefficient = standard distance. Subsequently, the binocular camera measurement distance of the object to be measured can be calibrated by using the standard coefficient to obtain the actual distance of the object to be measured, that is, the actual distance can be obtained by multiplying the binocular camera measurement distance of the object to be measured by the calibration coefficient.

[0047] In an embodiment of the present application, the steps of calibrating the binocular camera measurement distance based on the image coordinates of the object to be measured, the image coordinates of the calibration point, and the standard distance to obtain the actual distance include:

[0048] Calibrate the binocular camera measurement distance based on the image coordinates of the central point pixel of the object to be measured, the image coordinates of the calibration point, and the standard distance to obtain the actual distance of the object to be measured.

[0049] It can be understood that determining the distance of the object to be measured based on the central pixel point of the object to be measured can improve the accuracy of distance measurement.

[0050] In an embodiment of the present application, before the step of using the binocular camera 2 to obtain the light reflected by the object in the vehicle traveling direction to obtain a real-time image, the method further includes:

[0051] Controlling the single-point lidar 31 to irradiate the target object corresponding to the fixed pixel point in the image obtained by the binocular camera 2;

[0052] Determining the emission and reception duration of the single-point lidar 31 as the standard duration; determining the distance of the target object as the standard distance.

[0053] It can be understood that before the actual application of the vehicle ranging method, a calibration operation is first performed to obtain the standard duration and the standard distance. Specifically, controlling the single-point lidar 31 to irradiate the target object corresponding to the fixed pixel point in the image obtained by the binocular camera 2, obtaining the emission and reception duration of the single-point lidar 31, and determining the emission and reception duration of the single-point lidar 31 as the standard duration. At this time, the distance of the target object is determined, and the distance of the target object is defined as the standard distance, thereby correlating the standard duration and the standard distance.

[0054] In an embodiment of the present application, before the step of using the binocular camera 2 to obtain the light reflected by the object in the vehicle traveling direction to obtain a real-time image, the method further includes:

[0055] Determining the standard duration and the standard distance based on the actual tilt angle of the vehicle and the database;

[0056] The database includes the emission and reception duration of the single-point lidar 31 when irradiating the object corresponding to the fixed pixel point in the image obtained by the binocular camera 2 when the vehicle is at different tilt angles, and the distance of the object when the vehicle is at different tilt angles.

[0057] It can be understood that before the actual application of the vehicle ranging method, a calibration operation is first performed. Specifically, when the vehicle is at different angles, controlling the single-point lidar 31 to irradiate the object corresponding to the fixed pixel point in the image obtained by the binocular camera 2, obtaining the emission and reception duration of the single-point lidar 31, and determining the emission and reception duration of the single-point lidar 31 as the calibration duration, and determining the distance of the object as the calibration distance. That is to say, the calibration duration and the calibration distance corresponding to different angles of the vehicle will be obtained and the data will be stored in the database.

[0058] Before the step of obtaining the light reflected by the object in the traveling direction of the vehicle by using the binocular camera 2 to obtain a real-time image, obtain the actual tilt angle of the vehicle (which can also be understood as the tilt angle of the binocular camera 2), compare the actual tilt angle of the vehicle with the data in the database, output the calibration duration and calibration distance corresponding to the actual tilt angle of the vehicle, determine the corresponding calibration duration as the standard duration, and determine the corresponding calibration distance as the standard distance, so as to facilitate subsequent calibration of the distance of the object to be measured by using the standard duration and standard distance to obtain an accurate distance value. That is, in this embodiment, by obtaining the actual tilt angle of the vehicle and using the calibration duration and calibration distance corresponding to the actual tilt angle of the vehicle as the standard duration and standard distance, the accuracy of subsequent calibration of the distance of the object to be measured is improved.

[0059] It can be understood that when the vehicle is at different tilt angles, the pixels of the binocular camera 2 will be deformed to different degrees, and thus the standard duration and standard distance corresponding to different tilt angles of the vehicle are different.

[0060] In an embodiment of the present application, the actual distance is used to adjust the traveling strategy of the vehicle.

[0061] It can be understood that in a timely manner according to the actual distance of the object to be measured recognized, the processor can make corresponding decisions and actions. For example, avoiding the target vehicle, adjusting the path or stopping the traveling, etc.

[0062] In some embodiments, the traveling direction can be adjusted according to a preset route, and the preset route is determined based on the route calculated from the destination input by the user or the map data based on the current traveling direction, and there may be more than one preset route. Since operations such as turning and lane changing may be performed during the traveling process, the traveling direction is not limited to the direct front of the object and can be the area within a preset range near the traveling direction and the traveling direction.

[0063] When the object in front is a vehicle, in order to realize the detection of the vehicle, the vehicle ranging method further includes:

[0064] Step 101: Use a multi-channel spectral chip to obtain the light of different wavelengths reflected by the object in the traveling direction to obtain a real-time multi-spectral image, wherein the central wavelengths corresponding to different channels are different, so that different channels obtain spectral information of different wavelengths;

[0065] It can be understood that the multi-channel spectral chip can obtain the natural light reflected by the object. It can also be that different wavelengths of light are first emitted to the object by a light source, and the multi-channel spectral chip obtains the light emitted by the light source reflected by the object.

[0066] It can be understood that after the multi-channel spectral chip obtains the light rays of different wavelengths reflected by an object, a multi-spectral image in the traveling direction can be obtained based on the light rays of different wavelengths.

[0067] It can be understood that the multi-channel spectral chip has multiple channels, and the central wavelength corresponding to each channel is different, that is, the wavelength range of the light allowed to pass through each channel is different, so that the multi-channel spectral chip can obtain light rays of different wavelengths.

[0068] It can be understood that the traveling direction can include the front-back direction, and the front-back direction can exceed a range of 180 degrees.

[0069] Step 102: Process the real-time multi-spectral image to obtain tire features;

[0070] It can be understood that by processing the real-time multi-spectral image, it is judged whether there are tire features in the real-time multi-spectral image, so as to realize the extraction of the tire features in the real-time multi-spectral image.

[0071] It can be understood that the reflectivity of different materials is different at different wavelengths. In multi-spectral imaging, more accurate object recognition can be achieved by using the brightness features at different wavelengths. That is to say, in this application, the multi-channel spectral chip is used to obtain the light rays of different wavelengths reflected by the object in the traveling direction. Since different objects have different reflectivities for different wavelengths, that is, the spectral information carried by the light rays reflected by different objects is different, it is then possible to determine whether the object in the traveling direction is a tire according to the spectral information of different wavelengths obtained by different channels.

[0072] Step 103: Determine the vehicle features according to the tire features.

[0073] It can be understood that when it is determined that there are tire features in the real-time multi-spectral image, the vehicle features can be determined according to the tire features, that is, it can be determined that there is a vehicle in the real-time multi-spectral image according to the tire features, that is, it can be determined that there is a vehicle in the traveling direction.

[0074] It can be understood that each vehicle has tires, and the materials of the tires of different vehicles are basically the same. Therefore, by identifying the tires, the identification of the vehicle can be realized. Specifically, the vehicle features in the real-time multi-spectral image can be directly determined according to the tire features. It is also possible to further process the tire features according to the tire features, such as determining information such as the number and size of the tire features, and then determining whether the tire features can be used to represent the vehicle features.

[0075] According to the vehicle recognition method of the embodiments of the present application, a real-time multi-spectral image in the traveling direction is obtained through a multi-channel spectral chip, and the real-time multi-spectral image is processed to obtain the tires in the real-time multi-spectral image. When there are tire features in the real-time multi-spectral image, the vehicle features can be determined using the tire features. When there are no tire features in the real-time multi-spectral image, it means that there are no vehicle features in the real-time multi-spectral image. That is, the present application realizes the determination of the vehicle by determining the tire features, realizes the recognition of the vehicle using the tire features, and makes the vehicle recognition simpler. Moreover, the tire is a significant and essential feature of the vehicle, and using the tire features to determine the vehicle features makes the recognition of the vehicle features more stable and reliable.

[0076] In the vehicle feature recognition in the related art, mainly a large number of RGB images of real road conditions are collected through an optical camera, and a deep learning method is used for training to obtain a deep neural network model. This method requires a large number of pictures and obtains a complex recognition model after a long training time. The number of parameters is very large, and the calculation speed is limited. If real-time requirements are to be met, it poses a very high requirement for the performance of the computing device. However, the present application uses spectral information to identify and determine the tire features, determines the vehicle features through the tire features, and the recognition process is simpler, the recognition speed is fast, and it can meet the real-time requirements.

[0077] In an embodiment of the present application, processing the real-time multi-spectral image to obtain tire features includes:

[0078] Inputting the real-time multi-spectral image into a trained spectral model, where the spectral model is a neural network model, and the real-time multi-spectral image includes spectral information;

[0079] The spectral model calculates the real-time multi-spectral image and extracts the tire features.

[0080] It can be understood that when processing the real-time multi-spectral image, the real-time multi-spectral image can be input into the trained spectral model. The spectral model will process the spectral information carried by the real-time multi-spectral image. According to the spectral information carried by the real-time multi-spectral image, it can be determined whether there are tire features in the real-time multi-spectral image. When the spectral model determines that there are tire features in the real-time multi-spectral image according to the calculation result, the tire features will be extracted, that is, when the real-time multi-spectral image is input into the spectral model, the spectral image will output the tire features.

[0081] It can be understood that compared with the deep neural network model in the related art, the spectral model of the present application can determine whether there are tire features in the real-time multi-spectral image by processing the spectral information carried by the real-time multi-spectral image. The model is simpler, can respond quickly, and since the tire is an essential feature of the vehicle, the versatility of the spectral model is ensured.

[0082] In an embodiment of the present application, before inputting the real-time multispectral image into the trained spectral model, the vehicle recognition method further includes training the spectral model. Specifically, training the spectral model includes:

[0083] Obtaining a plurality of multispectral sample pictures;

[0084] Dividing each multispectral sample picture into a tire area and a non-tire area;

[0085] Obtaining the spectral information carried by the tire area and the non-tire area of each multispectral sample picture;

[0086] Establishing a spectral model based on the spectral information of the tire areas and non-tire areas of the plurality of multispectral sample pictures.

[0087] In an embodiment of the present application, processing the real-time multispectral image to obtain tire features includes:

[0088] Comparing the real-time multispectral image with the sample images in the spectral image library;

[0089] When there is a sample image in the spectral image library whose carried spectral information is the same as or similar to the spectral information carried by the real-time multispectral image, it is determined that there are tire features in the real-time multispectral image.

[0090] It can be understood that there are multiple sample images in the spectral image library, and each sample image has tire features. Since the scenes corresponding to different sample images will be different, the spectral information carried by different sample images will also be different.

[0091] When processing the real-time multispectral image, comparing the real-time multispectral image with the sample images in the spectral image library. When the spectral information of one of the sample images in the spectral image library is the same as or similar to the spectral information carried by the real-time multispectral image, it means that the real-time multispectral image has the same tire features as the sample image, and thus it can be determined that there are tire features in the real-time multispectral image.

[0092] Exemplarily, since the spectral information of the sample image is the same as or similar to the spectral information carried by the real-time multispectral image, it means that the scenes of the sample image and the real-time multispectral image are also the same or similar. Then, the tire features can be directly output based on the sample image. It is also possible to further identify the real-time multispectral image when it is determined that there are tire features in the real-time multispectral image, and identify the tire features in the real-time multispectral image.

[0093] In an embodiment of the present application, processing the real-time multispectral image to obtain tire features includes:

[0094] Compare the spectral information carried by the real-time multispectral image with the samples in the tire spectral information database;

[0095] When the spectral information carried by one object or area in the real-time multispectral image is the same as or similar to the spectral information of one tire sample in the tire spectral information database, it is determined that there is a tire feature in the real-time multispectral image.

[0096] It can be understood that there are multiple tire spectral information samples in the tire spectral information database, and each tire spectral information sample corresponds to a different scenario. That is, the tire spectral information database is established using the spectral information of tires in different scenarios.

[0097] When processing the real-time multispectral image, compare the spectral information carried by the real-time multispectral image with the samples in the tire spectral information database. When the spectral information corresponding to one object or one area in the real-time multispectral image is the same as or similar to the spectral information of one sample in the tire spectral information database, it indicates that one object or one area in the real-time multispectral image is a tire, and thus it can be determined that there is a tire feature in the real-time multispectral image.

[0098] Specifically, the objects in the real-time multispectral image can be identified and distinguished. The spectral information carried by different objects is different. Compare the spectral information carried by the object with the samples in the tire spectral information database. When the spectral information carried by the object is the same as or similar to the spectral information of one sample in the tire spectral information database, it indicates that the object is a tire, and thus it can be determined that there is a tire feature in the real-time multispectral image. It is also possible to divide the real-time multispectral image into multiple regions. The spectral information carried by each region is different. Compare the spectral information carried by each region with the samples in the tire spectral information database. When the spectral information carried by the region is the same as or similar to the spectral information of one sample in the tire spectral information database, it indicates that the region corresponds to a tire, and thus it can be determined that there is a tire feature in the real-time multispectral image.

[0099] In some examples, there are multiple objects or areas in the real-time multispectral image. Compare the spectral information carried by each object or each area with the samples in the tire spectral information database one by one. When the spectral information carried by one object or one area is the same as or similar to the spectral information of one sample in the tire spectral information database, it indicates that there is a tire feature in the real-time multispectral image, and then stop comparing the remaining objects or areas in the real-time multispectral image with the tire spectral information database.

[0100] In some examples, a real-time multi-spectral image has multiple objects or regions. The spectral information carried by each object or each region is compared one by one with the samples in the tire spectral information database until the comparison of the spectral information carried by all objects or all regions with the samples in the tire spectral information database is completed. Furthermore, it can be determined whether there are tire features in the real-time multi-spectral image and the number of tire features.

[0101] In an embodiment of the present application, based on the position of the tire feature in the real-time multi-spectral image, the position information of the target vehicle is determined.

[0102] It can be understood that according to the position of the tire feature in the real-time multi-spectral image, the position information of the tire can be determined, and further the position information of the target vehicle with the tire can be determined.

[0103] In an embodiment of the present application, the position information of the target vehicle is used to adjust the travel strategy.

[0104] It can be understood that in a timely manner according to the recognized position information of the target vehicle, the processor can make corresponding decisions and actions. For example, avoiding the target vehicle, adjusting the path, or stopping the travel, etc.

[0105] In some embodiments, the travel direction can be adjusted according to a preset route. The preset route is determined based on the route calculated from the destination input by the user or the map data based on the current travel direction. There may be more than one preset route. Since operations such as turning and lane changing may be performed during the travel, the travel direction is not limited to the exact front of the object and can be the travel direction and the area within a preset range near the travel direction.

[0106] In an embodiment of the present application, after processing the real-time multi-spectral image to obtain the tire feature, the vehicle recognition method further includes:

[0107] Obtain the spectral information of the tire feature;

[0108] Process the spectral information of the tire feature to obtain the wear degree of the tire.

[0109] It can be understood that by processing the spectral information of the tire feature, the wear degree of the tire can be determined, and further the travel strategy can be determined according to the wear degree of the tire. For example, when the wear of the tire exceeds the preset range, the distance from the tire feature can be controlled to increase.

[0110] Exemplarily, processing the spectral information of the tire feature includes:

[0111] Compare the spectral information of the tire feature with the tire wear spectral information database;

[0112] Determine a target sample from the samples in the tire wear spectrum information database that is the same as or similar to the spectrum information of the tire characteristics;

[0113] Take the wear degree corresponding to the target sample as the wear degree of the tire characteristics.

[0114] In some embodiments, after processing the real-time multi-spectral image to obtain the tire characteristics, the vehicle recognition method further includes:

[0115] Divide the tire characteristics into multiple secondary characteristics;

[0116] Process the spectrum information of the secondary characteristics to determine whether one or more of the secondary characteristics are made of metal.

[0117] It can be understood that the tire characteristics are divided into multiple secondary characteristics, and the spectrum information of each secondary characteristic is processed to determine whether one or more of the secondary characteristics are made of metal. When it is determined that the secondary characteristic is made of metal, it means that the tire carries a metal object, and then it is determined that the tire has a nail puncture.

[0118] In some embodiments, before the step of using the multi-channel spectral chip to obtain the light of different wavelengths reflected by the object in the traveling direction, it includes:

[0119] When the ambient light intensity is lower than the threshold, emit a light source signal in the traveling direction and irradiate the object in the traveling direction, and the light source signal includes light of multiple wavelengths.

[0120] It can be understood that when the ambient light intensity is lower than the threshold, it means that the ambient light is insufficient at this time. In order to ensure that the multi-channel spectral chip can obtain a clear multi-spectral image, first control the light source signal to emit light of multiple wavelengths to the object in the traveling direction, so that the multi-channel spectral chip can fully obtain the light of different wavelengths reflected by the object in the traveling direction.

[0121] It can be understood that when the ambient light intensity is greater than or equal to the threshold, it means that the ambient light is sufficient at this time, and the multi-channel spectral chip can also obtain a real-time multi-spectral image by obtaining the ambient light reflected by the object, then control the light source signal to turn off.

[0122] In order to further improve the intelligent degree of vehicle control, it is also possible to detect the material of the object while measuring the distance of the vehicle, so as to facilitate the adjustment of the traveling strategy.

[0123] Specifically, the vehicle distance measurement method further includes:

[0124] Emit a light source signal in the traveling direction and irradiate the object in the traveling direction, and the light source signal includes light of multiple wavelengths;

[0125] The multi-channel spectral chip is used to obtain light rays of different wavelengths reflected by an object to obtain spectral information of each channel. Among them, different channels correspond to spectral information of different wavelengths, and the light source signal includes light rays of the wavelengths corresponding to each channel;

[0126] The spectral information of multiple channels is processed to obtain the material parameters of the object.

[0127] Specifically, due to the diffuse reflection of the sky, white objects under cloudy skies will often have a brightness very similar to that of the sky, and traditional imaging techniques are prone to misidentification. Since solid objects do not emit or reflect light similar to that in the sky in the combination of spectral bands, a multispectral imaging system sensitive to spectra outside the visible light can more reliably determine the composition of the scene content. Therefore, the accuracy of the multispectral imaging technique is relatively high.

[0128] The reflectivities of different types of road surfaces are different at different wavelengths. The brightness characteristics at different wavelengths can be used in multispectral imaging to achieve more accurate object recognition. Different types of roads, vegetation, etc., such as snow, mud, weeds, and soil, can be detected and distinguished through multispectral imaging. Therefore, the ground type can be identified by analyzing these spectral reflection characteristics.

[0129] In some embodiments, after the spectral information of multiple channels is processed to obtain the material parameters of the object, it includes:

[0130] Adjust the traveling strategy according to the material parameters of the object.

[0131] It can be understood that different objects have different shapes and materials. In a timely manner based on the identified obstacles, the processor 13 can make corresponding decisions and actions. For example, avoiding obstacles, adjusting the path, or stopping the advancement, etc. The vehicle recognition system 10 further includes a controller. By processing the spectral information of multiple channels and feeding it back to the controller, the controller can more intelligently cope with different types of obstacles, improving the traveling efficiency and safety.

[0132] In some embodiments, the traveling direction can be adjusted according to a preset route. The preset route is determined based on the route calculated from the destination input by the user or the map data based on the current traveling direction. There may be more than one preset route. Since operations such as turning and lane changing may be performed during the traveling process, the traveling direction is not limited to the direct front of the object and can be the area within a preset range near the traveling direction and the traveling direction.

[0133] In some embodiments, adjusting the traveling strategy according to the material parameters of the object includes:

[0134] Determine whether the object is on the traveling route;

[0135] If an object is on the travel route, determine whether the object belongs to an obstacle according to the material parameters of the object;

[0136] If the object belongs to an obstacle, change the travel route to bypass the object.

[0137] Specifically, once an obstacle is recognized, the system will re-plan the driving route to avoid the obstacle. Considering the position, size and surrounding environment of the obstacle, the optimal driving path is selected. It can be understood that when the light source signal irradiates multiple objects in the traveling direction, the system can simultaneously process the spectral information of multiple objects. For example, if there is an obstacle in the front and no obstacle on both sides, either side can be selected to travel; if there is an obstacle in the front and another obstacle on one side, the side without an obstacle is selected to travel.

[0138] In some embodiments, determining whether an object belongs to an obstacle according to the material parameters of the object includes:

[0139] Obtain an image of the object;

[0140] Determine whether it belongs to an obstacle according to the image of the object and the material parameters of the object.

[0141] Specifically, in some cases, the material parameters of the object cannot determine the impact of the object on the road travel. For example, in the snow, small snow accumulations may reduce the travel speed but are not sufficient to hinder the travel. However, in the case of large snowdrifts on the road surface, it will not cause a great hindrance to the travel. The same is true on muddy ground. Therefore, it is necessary to further judge the object by combining the image of the object.

[0142] In some embodiments, an object detection algorithm can be used to identify obstacles in the traveling direction. The object detection algorithm can detect information such as the position, size and shape of different types of objects according to the image of the object. When traveling, the system can continuously obtain image data in the traveling direction and use the object detection model to determine whether the object belongs to an obstacle.

[0143] In some embodiments, adjusting the travel strategy according to the material parameters of the object includes:

[0144] Determine whether the object is the traveling ground;

[0145] If the object is the traveling ground, determine the ground type according to the material parameters of the object;

[0146] Adjust the travel speed according to the ground type.

[0147] Specifically, by analyzing and processing the spectral data, the processor 13 can identify the type of ground on which the current vehicle is located. If the object is the driving ground, it is necessary to further adjust the driving speed according to the ground type. The ground type includes different types of ground, such as vegetation, snow, etc., and can also be the ground type with different road surface conditions, such as slippery ground, potholed ground, etc. It can be understood that on a slippery muddy ground, the vehicle may need to reduce the speed to avoid slipping or losing control; while on a flat dry road, the vehicle can increase the speed to improve efficiency. If the material of the object identified ahead is not the driving road surface, the driving route is changed to bypass the object or the vehicle stops moving forward.

[0148] In some embodiments, adjusting the traveling strategy according to the material parameters of the object includes:

[0149] Confirming whether the object is a living object according to the material parameters of the object;

[0150] If the object is a living object, the distance from the object is monitored in real time, and the traveling speed is adjusted according to the distance.

[0151] Specifically, before confirming whether the object is a living object according to the material parameters of the object, it is necessary to collect image data containing living bodies and other environmental information and perform marking. Living bodies can be pedestrians, animals, etc. According to the detected living body information, in addition to the appearance characteristics, biometric recognition technologies, such as pedestrian gait recognition, animal movement patterns, etc., can also be used to assist in judging the living body. If the object is a living object, the processor 13 will monitor the distance from the object in real time and be able to make corresponding decisions and controls to adjust the traveling strategy, such as taking measures such as decelerating and avoiding. When the distance is less than the first threshold, the controller decelerates and changes lanes, and when the distance is less than the second threshold, the controller controls the vehicle to stop.

[0152] In some embodiments, processing the spectral information of multiple channels to obtain the material parameters of the object includes:

[0153] Collecting the spectral information to be measured of multiple channels to form sample spectral information, comparing the sample spectral information with the samples in the spectral library, and determining the samples in the target spectral library that are the same as or similar to the sample spectral information from the samples in the spectral library;

[0154] Taking the object material parameters corresponding to the target samples in the spectral library as the object material parameters.

[0155] In some embodiments, the spectral information is data information. Before collecting the spectral information to be measured of multiple channels to form sample spectral information and comparing the sample spectral information with the samples in the spectral library, preprocessing can be performed on the multiple spectral information collected for one object, including operations such as noise removal and data correction, to obtain a set of effective sample spectral information.

[0156] In some embodiments, curve fitting can be performed on the preprocessed spectral information. The curve obtained after fitting can be used to describe the overall characteristics of the spectral information, and sample spectral features can be extracted from the fitted curve. These features can include information such as spectral peak position, spectral peak intensity, and spectral bandwidth, which are used to describe the local features and structure of the spectral information. By comparing the sample spectral features with the spectral features in the spectral library, if the local features and structure of the spectral information are the same or similar, the object material parameters corresponding to the target sample in the spectral library can be regarded as the material parameters of the object.

[0157] In some examples, the mean square error can be calculated between the sample spectral information and the spectral data in the spectral library to obtain multiple sets of data, and the object material parameters of the target sample corresponding to the minimum mean square error in the data can be used as the material parameters of the object.

[0158] In some embodiments, the spectral information is image information. Before forming the sample spectral information by aggregating the spectral information to be measured in multiple channels and comparing the sample spectral information with the samples in the spectral library, preprocessing of the multi-spectral image can be performed, including operations such as noise removal, image correction, and contrast enhancement, to obtain a set of effective sample spectral information.

[0159] In some embodiments, a multi-spectral image can be obtained through a multi-channel spectral chip, and the required feature data can be extracted, such as color histograms, texture features, shape features, etc. By comparing the sample spectral image with the spectral images in the spectral library and determining the target spectral image that is the same or similar to the sample spectral image from the spectral images in the spectral library, the object material parameters corresponding to the target image in the spectral library can be regarded as the material parameters of the object.

[0160] In some embodiments, processing the spectral information of multiple channels to obtain the material parameters of the object includes:

[0161] Aggregating the spectral information to be measured in multiple channels to form sample spectral information, and inputting the sample spectral information into a spectral model, where the spectral model is a neural network model;

[0162] The spectral model calculates the sample spectral information and generates the material parameters of the object.

[0163] Specifically, before inputting the sample spectral information into the spectral model, the original model can be trained by inputting a large amount of training data, where the training data includes the corresponding spectral information and material parameters, to obtain the trained model, and the trained model is the spectral model.

[0164] Assume that the material parameters of each object include several different spectral data. Then, each object in the spectral model also includes several different spectral data. The new spectral data is input into the model, and the big data is operated with the data in the spectral model to obtain data that matches the current object material model.

[0165] According to an embodiment of the second aspect of the present application, as Figure 2 , the vehicle ranging device and the vehicle ranging method are correspondingly referred to each other. As Figure 3 shown, the vehicle ranging device includes:

[0166] A first acquisition module 201, configured to use a binocular camera 2 to acquire the light reflected by an object in the traveling direction of the vehicle to obtain a real-time image;

[0167] A second acquisition module 202, configured to acquire the emission and reception duration of multiple single-point lidars 31, where different single-point lidars 31 correspond to the objects corresponding to different pixel points of the real-time image;

[0168] A first determination module 203, configured to determine the object irradiated by the target single-point lidar 31 as a calibration point, where the emission and reception duration of the target single-point lidar 31 is equal to the standard duration;

[0169] A second determination module 204, configured to determine the binocular camera measurement distance of the object to be measured in the real-time image;

[0170] A control module 205, configured to calibrate the binocular camera measurement distance based on the image coordinates of the object to be measured, the image coordinates of the calibration point, and the standard distance to obtain the actual distance;

[0171] Wherein, the emission and reception duration of the single-point lidar 31 irradiating the object at the standard distance is the standard duration.

[0172] According to an embodiment of the third aspect of the present application, as Figure 3 , the vehicle ranging system includes:

[0173] A vehicle body 1;

[0174] At least two binocular cameras 2, provided on the front hood 11 of the vehicle body 1, and the binocular cameras 2 are configured to acquire real-time images in the traveling direction of the vehicle body 1;

[0175] At least two single-point laser modules 3, provided on the front hood 11 of the vehicle body 1, at least two single-point laser modules 3 are arranged in one-to-one correspondence with at least two binocular cameras 2, the single-point laser module 3 is located between the binocular camera 2 and the front windshield of the vehicle body 1, and the single-point laser module 3 is configured to irradiate the object corresponding to the fixed pixel point of the real-time image to detect the distance of the object.

[0176] It can be understood that a single-point laser module 3 is correspondingly arranged at each binocular camera 2, and precise ranging of the objects in front of the vehicle body 1 is achieved through the combination of the binocular camera 2 and the single-point laser module 3. Moreover, by arranging the binocular camera 2 and the single-point laser module 3 on the front hood 11 of the vehicle, the shooting angle is kept consistent with the vehicle driving angle, which can better detect the objects in front. At the same time, by arranging at least two binocular cameras 2 and the corresponding single-point laser modules 3, the shooting interface is expanded, the detection range is expanded, and the ranging effect can be effectively improved.

[0177] In an embodiment of the present application, as Figure 4 , the single-point laser module 3 includes at least three rows of single-point lidars 31, and the detection distances of the single-point lidars 31 in different rows are different.

[0178] It can be understood that by arranging at least three rows of single-point lidars 31, each row of lidars includes at least two single-point lidars 31, and each row of single-point lidars 31 correspondingly detects different distances, which can avoid the problems of decreased detection accuracy or increased data processing difficulty caused by using one single-point lidar 31 to detect multiple distances.

[0179] In an embodiment of the present application, the two binocular cameras 2 are symmetrically arranged about the midline of the vehicle body 1, and the two single-point laser modules 3 are symmetrically arranged about the midline of the vehicle body 1.

[0180] It can be understood that one of the binocular cameras 2 and the corresponding single-point laser module 3 can range the objects on one side of the vehicle body 1, and the other binocular camera 2 and the corresponding single-point laser module 3 can range the objects on the other side of the vehicle body 1.

[0181] The two binocular cameras 2 are symmetrically arranged, and the two single-point laser modules 3 are symmetrically arranged, so that the two cameras are at the same height, and the two single-point laser modules 3 are at the same height, ensuring that the lower edges of the images captured by the two binocular cameras 2 are at the same height, avoiding the situation where the lower edges of the images captured by the two binocular cameras 2 are at different heights, and thus avoiding the occurrence of shooting blind spots.

[0182] In an embodiment of the present application, as Figure 4 , the single-point laser module 3 includes multiple single-point lidars 31, and different single-point lidars 31 are used to irradiate the objects corresponding to different pixel points of the real-time image.

[0183] It can be understood that by arranging multiple single-point lidars 31, the objects corresponding to different pixel points can be ranged.

[0184] In one embodiment of the present application, the binocular camera 2 includes a binocular camera base and a binocular camera body. The binocular camera base is detachably connected to the front hood 11 of the vehicle, and the binocular camera body is connected to the binocular camera base.

[0185] It can be understood that the binocular camera base is detachably connected to the front hood 11, that is, the binocular camera 2 is detachably connected to the front hood 11, which facilitates the disassembly and assembly of the binocular camera 2.

[0186] In one embodiment of the present application, along the direction from the binocular camera 2 to the single-point laser module 3, the cross-sectional area of the binocular camera base gradually increases so that the binocular camera base is attached to the front hood 11.

[0187] It can be understood that the front hood 11 of the vehicle body 1 gradually slopes downward along the direction from the single-point laser module 3 to the binocular camera 2. In this embodiment, the cross-sectional area of the binocular camera base is set to gradually increase along the direction from the binocular camera 2 to the single-point laser module 3, so that when the vehicle is at a horizontal angle, the binocular camera 2 is also horizontal, ensuring the shooting angle of the binocular camera 2.

[0188] In one embodiment of the present application, the single-point laser module 3 includes a single-point laser base and a single-point laser body. The single-point laser base is detachably connected to the front hood 11 of the vehicle, and the single-point laser body is connected to the single-point laser base.

[0189] It can be understood that the single-point laser base is detachably connected to the front hood 11, that is, the single-point laser module 3 is detachably connected to the front hood 11, which facilitates the disassembly and assembly of the single-point laser module 3.

[0190] In one embodiment of the present application, along the direction from the binocular camera 2 to the single-point laser module 3, the cross-sectional area of the single-point laser base gradually increases so that the single-point laser base is attached to the front hood 11.

[0191] It can be understood that the front hood 11 of the vehicle body 1 gradually slopes downward along the direction from the single-point laser module 3 to the binocular camera 2. In this embodiment, the cross-sectional area of the single-point laser base is set to gradually increase along the direction from the binocular camera 2 to the single-point laser module 3, so that when the vehicle is at a horizontal angle, the single-point laser module 3 is also horizontal, ensuring the shooting angle of the single-point laser module 3.

[0192] In one embodiment of the present application, the shooting area of one of the binocular cameras 2 is adjacent to or partially overlaps with the shooting area of the other binocular camera 2.

[0193] It can be understood that setting the shooting areas of the two binocular cameras 2 to be adjacent can maximize the shooting area and there is no shooting blind spot.

[0194] It can be understood that the shooting areas of the two binocular cameras 2 are set to partially overlap, and the objects in the overlapping part will be measured by both the two binocular cameras 2 and the corresponding single-point laser modules 3, ensuring the accuracy of the distance measurement of the objects in the overlapping part.

[0195] According to the embodiment of the fourth aspect of the present application, the vehicle includes the above vehicle distance measurement system.

[0196] For the vehicle according to the embodiment of the present application, a single-point laser module 3 is correspondingly arranged at each binocular camera 2, and the combination of the binocular camera 2 and the single-point laser module 3 is used to accurately measure the distance of the objects in front of the vehicle body 1. And by arranging the binocular camera 2 and the single-point laser module 3 on the front hood 11, the shooting angle is kept consistent with the vehicle driving angle, which can better detect the objects in front. At the same time, by arranging at least two binocular cameras 2 and the corresponding single-point laser modules 3, the shooting interface is expanded, the detection range is expanded, and the distance measurement effect can be effectively improved.

[0197] As Figure 5 shown, the electronic device may include: a processor 310, a communication interface 320, a memory 330, and a communication bus 340. Among them, the processor 310, the communication interface 320, and the memory 330 communicate with each other through the communication bus 340. The processor 310 can call the logical instructions in the memory 330 to execute the vehicle distance measurement method, and the method includes:

[0198] Utilize the binocular camera 2 to obtain the light reflected by the objects in the vehicle traveling direction to obtain a real-time image;

[0199] Obtain the emission and reception durations of multiple single-point lidars 31, and different single-point lidars 31 correspond to the objects corresponding to different pixel points of the real-time image;

[0200] Determine the object irradiated by the target single-point lidar 31 as the calibration point, where the emission and reception duration of the target single-point lidar 31 is equal to the standard duration;

[0201] Determine the binocular camera measurement distance of the object to be measured in the real-time image;

[0202] Based on the image coordinates of the object to be measured, the image coordinates of the calibration point, and the standard distance, calibrate the binocular camera measurement distance to obtain the actual distance;

[0203] Wherein, the emission and reception duration of the single-point lidar 31 irradiating the object at the standard distance is the standard duration.

[0204] In addition, when the logical instructions in the above-mentioned memory 330 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0205] On the other hand, the present application also provides a computer program product. The computer program product includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the vehicle ranging method provided by the above-mentioned various methods. The method includes:

[0206] Utilize the binocular camera 2 to obtain the light reflected by the objects in the vehicle traveling direction to obtain a real-time image;

[0207] Obtain the emission and reception durations of multiple single-point lidars 31. Different single-point lidars 31 correspond to the objects corresponding to different pixel points of the real-time image irradiated;

[0208] Determine the object irradiated by the target single-point lidar 31 as the calibration point, where the emission and reception duration of the target single-point lidar 31 is equal to the standard duration;

[0209] Determine the binocular camera measurement distance of the object to be measured in the real-time image;

[0210] Based on the image coordinates of the object to be measured, the image coordinates of the calibration point, and the standard distance, calibrate the binocular camera measurement distance to obtain the actual distance;

[0211] Wherein, the emission and reception duration of the single-point lidar 31 irradiating an object at the standard distance is the standard duration.

[0212] According to the embodiments of the fifth aspect of the present application, the present application further includes a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is implemented to execute the vehicle ranging method provided by the above-mentioned various methods. The method includes:

[0213] Utilize the binocular camera 2 to obtain the light reflected by the objects in the vehicle traveling direction to obtain a real-time image;

[0214] Obtain the emission and reception durations of multiple single-point lidars 31, where different single-point lidars 31 correspond to objects corresponding to different pixel points of the live image being irradiated;

[0215] Determine the object irradiated by the target single-point lidar 31 as the calibration point, where the emission and reception duration of the target single-point lidar 31 is equal to the standard duration;

[0216] Determine the binocular camera measurement distance of the object to be measured in the live image;

[0217] Based on the image coordinates of the object to be measured, the image coordinates of the calibration point, and the standard distance, calibrate the binocular camera measurement distance to obtain the actual distance;

[0218] Wherein, the emission and reception duration of the single-point lidar 31 irradiating an object at the standard distance is the standard duration.

[0219] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without creative labor.

[0220] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course also by hardware. Based on this understanding, the above technical solution, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiments.

[0221] The above has introduced the embodiments of the present application in detail. Specific examples are used in this article to elaborate on the principle and implementation manner of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application; at the same time, for those skilled in the art, according to the idea of the present application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A vehicle distance measurement system, characterized in that: include: Vehicle body; At least two binocular cameras are arranged on the front hood of the vehicle body, and the binocular cameras are used to obtain real-time images in the traveling direction of the vehicle body; At least two single-point laser modules are arranged on the front hood of the vehicle body. The at least two single-point laser modules are arranged one by one opposite to the at least two binocular cameras. The single-point laser modules are located between the binocular cameras and the windshield of the vehicle body. The single-point laser modules are used to illuminate the object corresponding to the fixed pixel point of the real-time image to detect the distance of the object.

2. The vehicle distance measurement system according to claim 1, characterized in that: The single-point laser module includes at least three rows of single-point laser radars, and the detection distances of the single-point laser radars in different rows are different.

3. The vehicle distance measurement system according to claim 1, characterized in that: The two binocular cameras are symmetrically arranged about the center line of the vehicle body, and the two single-point laser modules are symmetrically arranged about the center line of the vehicle body.

4. The vehicle distance measurement system according to claim 1, characterized in that: The single-point laser module includes a plurality of single-point laser radars, and different single-point laser radars are used to illuminate objects corresponding to different pixel points aligned with the real-time image.

5. The vehicle distance measurement system according to any one of claims 1 to 4, characterized in that: The binocular camera comprises a binocular camera base and a binocular camera body. The binocular camera base is detachably connected to the vehicle hood, and the binocular camera body is connected to the binocular camera base.

6. The vehicle distance measurement system according to claim 5, characterized in that: Along the direction from the binocular camera to the single-point laser module, the cross-sectional area of ​​the binocular camera base gradually increases, so that the binocular camera base is attached to the vehicle hood.

7. The vehicle distance measurement system according to any one of claims 1 to 4, characterized in that: The single-point laser module comprises a single-point laser base and a single-point laser body. The single-point laser base is detachably connected to the vehicle hood, and the single-point laser body is connected to the single-point laser base.

8. The vehicle distance measurement system according to claim 7, characterized in that: Along the direction from the binocular camera to the single-point laser module, the cross-sectional area of ​​the single-point laser base gradually increases, so that the single-point laser base is attached to the front cover of the vehicle.

9. The vehicle distance measurement system according to any one of claims 1 to 4, characterized in that: A shooting area of ​​one of the binocular cameras is adjacent to or partially overlaps with a shooting area of ​​another binocular camera.

10. A vehicle, characterized in that: The invention comprises a vehicle distance measurement system as claimed in any one of claims 1 to 9.

Images