Vehicle end roadside sensing method, system and device based on monocular vision

Abstract

The invention discloses a vehicle end roadside sensing method, system and device based on monocular vision, and the method, system and device are implemented, before lane sensing, through the analysis of first scene data and the judgment of set roadside sensing conditions, the sensing in an improper scene is avoided, and the improvement of the accuracy and reliability of subsequent operation is facilitated; then, the second scene data collected in real time is combined with the basic road information, so that the real-time monitoring of the relationship between the vehicle and the road boundary in the driving process is realized, and the real-time performance and the accuracy of the sensing result are improved; and finally, by accurately calculating the roadside proportion information, the position of the vehicle on the road can be monitored in real time, and visual vehicle position feedback is provided for a driver, so that the driver is helped to timely correct the driving direction of the vehicle and avoid deviating from the road or collision accidents, and the driving safety and the road driving stability are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of vehicle navigation and positioning technology, and in particular to a vehicle-side roadside perception method, system and device based on monocular vision. Background Art

[0002] Roadside perception is a technology that uses multiple sensor technologies and combines them with roadside edge computing to achieve instantaneous intelligent perception of traffic participants, road conditions and other information. It is also an important part of the field of smart transportation and autonomous driving.

[0003] In order to obtain more accurate roadside information in roadside perception technology, technicians have tried to apply different types of sensors to the perception system, such as binocular cameras, event cameras, monocular cameras, millimeter-wave radars, lidars and other sensors. Data processing technology and communication technology adapted to the perception system have also become the focus of research by technicians. However, from the perspective of practical application, the more types of sensors are used, the more complex the corresponding data processing technology, and the higher the configuration requirements for the communication network, resulting in a more complex system structure and higher costs, and the reliability of the system operation will be difficult to guarantee. Summary of the invention

[0004] The present invention provides a vehicle-side roadside perception method, system and device based on monocular vision, which can help reduce the computational complexity and amount of roadside perception, while improving the computational accuracy and efficiency of roadside perception.

[0005] In order to solve the above technical problems, the first aspect of the present invention discloses a vehicle-side roadside perception method based on monocular vision, the method comprising:

[0006] Collecting first scene data of the target vehicle in the current scene before it starts, the first scene data including first vehicle front data within a preset viewing angle in front of the target vehicle and first vehicle posture information of the target vehicle;

[0007] After determining that the first scene data indicates that a preset roadside perception condition is currently satisfied, analyzing the first scene data to obtain basic road information of the front scene, the basic road information at least including a road width and a target camera angle corresponding to a monocular camera device;

[0008] Real-time collection of second scene data of the current scene after the target vehicle starts to travel, the second scene data including second vehicle front data and second vehicle posture information;

[0009] According to a preset roadside perception algorithm, in combination with the basic road information and the second scene data, target measurement information is calculated, the target measurement information including roadside ratio information of the target vehicle and the road; wherein the roadside ratio information is used to indicate a deviation attribute of the target vehicle on the road, the deviation attribute including any one of a left deviation attribute indicating that the target vehicle is located on the left side of the road, a right deviation attribute indicating that the target vehicle is located on the right side of the road, and a centering attribute indicating that the target vehicle is located in the middle of the road.

[0010] As an optional implementation, in the first aspect of the present invention, two roadside signs are provided on both sides of the current scene, one roadside sign is provided on each side of the road, and the two roadside signs are provided on the same horizontal line, and the horizontal line is perpendicular to the road; the preset viewing angle range is the device photography range of the monocular camera device;

[0011] After collecting the first scene data of the current scene before the target vehicle is started, the method further includes:

[0012] Determining whether the first vehicle-front data indicates that the two roadside signs can be identified within the photographic range of the device, and when it is determined that the first vehicle-front data indicates that the two roadside signs can be identified within the photographic range of the device, determining that the first scene data indicates that a preset roadside perception condition is currently satisfied;

[0013] When it is determined that the first vehicle-front data indicates that the two roadside signs cannot be identified within the photographic range of the device, the position adjustment is performed on the target vehicle; and after determining that the monocular camera device is adjusted to be able to simultaneously identify the two roadside signs, the first scene data is updated, and it is determined that the first scene data indicates that the preset roadside perception conditions are currently met.

[0014] As an optional implementation, in the first aspect of the present invention, the first vehicle posture information includes a device position corresponding to a monocular camera device;

[0015] The analyzing the first scene data to obtain basic road information of the front scene includes:

[0016] According to a preset target recognition algorithm, taking the road side line as the recognition object, performing target recognition on the first vehicle front data to obtain the road width corresponding to the road in the current scene;

[0017] Reading the photographing range of the monocular camera device in the current scene and its corresponding photographing angle;

[0018] Taking the device position as a reference point and taking the direction parallel to the road side line as a reference direction, performing angle division on the photography angle, and obtaining the angle between the monocular camera device and the road side line on the left as the target photography angle;

[0019] The road width and the target camera angle are determined as basic road information of the front scene.

[0020] As an optional implementation manner, in the first aspect of the present invention, the calculating the target measurement information according to the preset roadside perception algorithm in combination with the basic road information and the second scene data includes:

[0021] The two roadside signs are divided into a first sign located on the left side of the road and a second sign located on the right side of the road; wherein, after the target vehicle starts to travel, the roadside sign that the target vehicle travels away from is the first sign, and the roadside sign that the target vehicle travels towards is the second sign;

[0022] Determine, based on the second vehicle front data, a time difference between the two roadside signs disappearing in sequence after the target vehicle starts to travel;

[0023] determining an average travel speed of the target vehicle according to the second vehicle posture information;

[0024] According to the time difference, the average driving speed, the road width, and a preset roadside ratio calculation formula, a target roadside ratio value is determined as target measurement information.

[0025] As an optional embodiment, in the first aspect of the present invention, the method further comprises:

[0026] Determining a disappearance order of the two roadside signs according to the target roadside ratio and the second vehicle front data;

[0027] Determining the deviation information of the target vehicle on the road according to the disappearance order of the two roadside signs;

[0028] When the deviation information indicates that the target vehicle is located in the middle of the road, a safety warning is generated for the target vehicle to prompt the driver of the target vehicle to make driving adjustments;

[0029] Among them, when the second data in front of the vehicle indicates that the first mark disappears first, the deviation information indicates that the target vehicle is located on the right side of the road; when the second data in front of the vehicle indicates that the second mark disappears first, the deviation information indicates that the target vehicle is located on the left side of the road; when the second data in front of the vehicle indicates that the first mark and the second mark disappear at the same time, the deviation information indicates that the target vehicle is located in the middle of the road.

[0030] As an optional implementation manner, in the first aspect of the present invention, the determination method corresponding to the roadside ratio calculation formula includes:

[0031] Determine the first node data and the second node data according to the second vehicle front data and the second vehicle posture information; the first node data includes a first recording time and a first vehicle position corresponding thereto; the second node data includes a second recording time and a second vehicle position corresponding thereto; the first recording time is a time corresponding to the initial disappearance of the first mark after the target vehicle starts to travel; the second recording time is a time corresponding to the initial disappearance of the second mark after the target vehicle starts to travel;

[0032] Establish a coordinate system with the first vehicle position as the origin, the direction corresponding to the road sideline as the longitudinal axis, and the direction perpendicular to the road sideline as the transverse axis; and determine the first coordinate, the second coordinate, and the third coordinate of the first identifier, the second identifier, and the second vehicle position in the coordinate system;

[0033] Determine the driving deviation angle of the target vehicle according to the third coordinate, wherein the driving deviation angle is the included angle between the longitudinal axis and the driving distance; the driving distance is the straight-line distance between the first vehicle position and the second vehicle position;

[0034] A roadside ratio calculation formula is determined according to the first coordinate, the second coordinate, the third coordinate, the driving distance, the driving deviation angle, and a set basic roadside ratio formula.

[0035] As an optional implementation, in the first aspect of the present invention, the formula corresponding to the roadside ratio calculation formula is specifically:

[0036]

[0037] Wherein, λ corresponds to the ratio between the target vehicle to the right side of the road and the target vehicle to the left side of the road; is the motion vector of the target vehicle, which includes the driving deviation angle ε and the average driving speed v of the target vehicle; t is the moving time corresponding to the average driving speed; S is the road width of the road, and θ is the target camera angle.

[0038] The second aspect of the present invention discloses a vehicle-side roadside perception system based on monocular vision, the system comprising:

[0039] A data acquisition module, used to collect first scene data of the target vehicle in the current scene before it is started, wherein the first scene data includes first vehicle front data within a preset viewing angle range in front of the target vehicle and first vehicle posture information of the target vehicle;

[0040] A data analysis module, configured to analyze the first scene data to obtain basic road information of the front scene after determining that the first scene data indicates that a preset roadside perception condition is currently satisfied, wherein the basic road information includes at least a road width and a target camera angle corresponding to a monocular camera device;

[0041] The data acquisition module is further used to collect in real time the second scene data of the current scene after the target vehicle starts to travel, wherein the second scene data includes the second vehicle front data and the second vehicle posture information;

[0042] A calculation module is used to calculate target measurement information according to a preset roadside perception algorithm, in combination with the basic road information and the second scene data, wherein the target measurement information includes roadside ratio information of the target vehicle and the road; wherein the roadside ratio information is used to indicate a deviation attribute of the target vehicle on the road, and the deviation attribute includes any one of a left deviation attribute indicating that the target vehicle is located on the left side of the road, a right deviation attribute indicating that the target vehicle is located on the right side of the road, and a centering attribute indicating that the target vehicle is located in the middle of the road.

[0043] As an optional implementation, in the second aspect of the present invention, two roadside signs are provided on both sides of the current scene, one roadside sign is provided on each side of the road, and the two roadside signs are provided on the same horizontal line, and the horizontal line is perpendicular to the road; the preset viewing angle range is the device photography range of the monocular camera device;

[0044] The system further comprises:

[0045] A judgment module, configured to judge whether the first vehicle front data indicates that the two roadside signs can be identified within the photographing range of the device after the data collection module collects the first scene data of the current scene before the target vehicle starts;

[0046] a determination module, configured to determine that the first scene data currently satisfies a preset roadside perception condition when the determination module determines that the first vehicle front data indicates that the two roadside signs can be identified within the photographing range of the device;

[0047] A posture adjustment module, configured to perform posture adjustment on the target vehicle when the judgment module judges that the first vehicle front data indicates that the two roadside signs cannot be identified within the photographic range of the device;

[0048] A data updating module is used to update the first scene data after determining that the monocular camera device is adjusted to be able to simultaneously identify the two roadside signs, and to determine that the first scene data indicates that a preset roadside perception condition is currently met.

[0049] As an optional implementation, in the second aspect of the present invention, the first vehicle posture information includes a device position corresponding to a monocular camera device;

[0050] The data analysis module analyzes the first scene data to obtain the basic road information of the front scene, which specifically includes:

[0051] According to a preset target recognition algorithm, taking the road side line as the recognition object, performing target recognition on the first vehicle front data to obtain the road width corresponding to the road in the current scene;

[0052] Reading the photographing range of the monocular camera device in the current scene and its corresponding photographing angle;

[0053] Taking the device position as a reference point and taking the direction parallel to the road side line as a reference direction, performing angle division on the photography angle, and obtaining the angle between the monocular camera device and the road side line on the left as the target photography angle;

[0054] The road width and the target camera angle are determined as basic road information of the front scene.

[0055] As an optional implementation, in the second aspect of the present invention, the calculation module calculates the target measurement information according to a preset roadside perception algorithm in combination with the basic road information and the second scene data, specifically including:

[0056] The two roadside signs are divided into a first sign located on the left side of the road and a second sign located on the right side of the road; wherein, after the target vehicle starts to travel, the roadside sign that the target vehicle travels away from is the first sign, and the roadside sign that the target vehicle travels towards is the second sign;

[0057] Determine, based on the second vehicle front data, a time difference between the two roadside signs disappearing in sequence after the target vehicle starts to travel;

[0058] determining an average travel speed of the target vehicle according to the second vehicle posture information;

[0059] According to the time difference, the average driving speed, the road width, and a preset roadside ratio calculation formula, a target roadside ratio value is determined as target measurement information.

[0060] As an optional implementation, in the second aspect of the present invention, the determination module is further used to determine the disappearance order of the two roadside signs according to the target roadside ratio and the second vehicle front data;

[0061] The determination module is further used to determine the deviation information of the target vehicle on the road according to the disappearance order of the two roadside signs;

[0062] The system further comprises:

[0063] A safety warning module, configured to generate a safety warning for the target vehicle when the deviation information indicates that the target vehicle is located in the middle of the road, so as to prompt the driver of the target vehicle to make driving adjustments;

[0064] Among them, when the second data in front of the vehicle indicates that the first mark disappears first, the deviation information indicates that the target vehicle is located on the right side of the road; when the second data in front of the vehicle indicates that the second mark disappears first, the deviation information indicates that the target vehicle is located on the left side of the road; when the second data in front of the vehicle indicates that the first mark and the second mark disappear at the same time, the deviation information indicates that the target vehicle is located in the middle of the road.

[0065] As an optional implementation, in the second aspect of the present invention, the determination method corresponding to the roadside ratio calculation formula specifically includes:

[0066] Determine the first node data and the second node data according to the second vehicle front data and the second vehicle posture information; the first node data includes a first recording time and a first vehicle position corresponding thereto; the second node data includes a second recording time and a second vehicle position corresponding thereto; the first recording time is a time corresponding to the initial disappearance of the first mark after the target vehicle starts to travel; the second recording time is a time corresponding to the initial disappearance of the second mark after the target vehicle starts to travel;

[0067] Establish a coordinate system with the first vehicle position as the origin, the direction corresponding to the road sideline as the longitudinal axis, and the direction perpendicular to the road sideline as the transverse axis; and determine the first coordinate, the second coordinate, and the third coordinate of the first identifier, the second identifier, and the second vehicle position in the coordinate system;

[0068] Determine the driving deviation angle of the target vehicle according to the third coordinate, wherein the driving deviation angle is the included angle between the longitudinal axis and the driving distance; the driving distance is the straight-line distance between the first vehicle position and the second vehicle position;

[0069] A roadside ratio calculation formula is determined according to the first coordinate, the second coordinate, the third coordinate, the driving distance, the driving deviation angle, and a set basic roadside ratio formula.

[0070] As an optional implementation, in the second aspect of the present invention, the formula corresponding to the roadside ratio calculation formula is specifically:

[0071]

[0072] Among them, λ corresponds to the ratio between the target vehicle to the right side of the road and the target vehicle to the left side of the road; →V is the motion vector of the target vehicle, which includes the driving deviation angle ε and the average driving speed v of the target vehicle; t is the moving time corresponding to the average driving speed; S is the road width of the road, and θ is the target camera angle.

[0073] The third aspect of the present invention discloses another vehicle-side roadside perception device based on monocular vision, the device comprising:

[0074] A memory storing executable program code;

[0075] a processor coupled to the memory;

[0076] The processor calls the executable program code stored in the memory to execute the vehicle-side roadside perception method based on monocular vision disclosed in the first aspect of the present invention.

[0077] The fourth aspect of the present invention discloses a computer storage medium, which stores computer instructions. When the computer instructions are called, they are used to execute the vehicle-side roadside perception method based on monocular vision disclosed in the first aspect of the present invention.

[0078] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0079] In an embodiment of the present invention, a vehicle-side roadside perception method based on monocular vision is provided, the method comprising: collecting first scene data of a target vehicle in a current scene before it starts, the first scene data comprising first vehicle front data within a preset viewing angle range in front of the target vehicle and first vehicle posture information of the target vehicle; after determining that the first scene data indicates that a preset roadside perception condition is currently satisfied, analyzing the first scene data to obtain basic road information of the front scene, the basic road information comprising at least a road width and a target camera angle corresponding to a monocular camera device; collecting second scene data of the target vehicle in the current scene in real time after it starts driving, the second scene data comprising second vehicle front data and second vehicle posture information; calculating target measurement information according to a preset roadside perception algorithm, combining the basic road information and the second scene data, the target measurement information comprising roadside ratio information of the target vehicle and the road; wherein the roadside ratio information is used to indicate a deviation attribute of the target vehicle on the road, the deviation attribute comprising any one of a left deviation attribute indicating that the target vehicle is located on the left side of the road, a right deviation attribute indicating that the target vehicle is located on the right side of the road, and a centering attribute indicating that the target vehicle is located in the middle of the road. It can be seen that by implementing the present invention, before lane perception, by analyzing the first scene data and judging the set roadside perception conditions, perception in inappropriate scenes is avoided, which is beneficial to improving the accuracy and reliability of subsequent operations; then, by combining the second scene data collected in real time with the basic road information, not only the real-time monitoring of the relationship between the vehicle and the road boundary during driving is realized, but also the real-time and accuracy of the perception results are improved; finally, by accurately calculating the roadside proportion information, the position of the vehicle on the road can be monitored in real time, and the driver can be provided with intuitive vehicle position feedback, which helps the driver to correct the vehicle's driving direction in time, avoid deviation from the road or collision accidents, thereby significantly improving driving safety and road driving stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0081] Figure 1 It is a flow chart of a vehicle-side roadside perception method based on monocular vision disclosed in an embodiment of the present invention;

[0082] Figure 2 It is a flow chart of another vehicle-side roadside perception method based on monocular vision disclosed in an embodiment of the present invention;

[0083] Figure 3It is a structural schematic diagram of a vehicle-side roadside perception system based on monocular vision disclosed in an embodiment of the present invention;

[0084] Figure 4 It is a structural schematic diagram of another vehicle-side roadside perception system based on monocular vision disclosed in an embodiment of the present invention;

[0085] Figure 5 It is a structural schematic diagram of a vehicle-side roadside perception device based on monocular vision disclosed in an embodiment of the present invention;

[0086] Figure 6 It is a working schematic diagram of a vehicle-side roadside perception system based on monocular vision disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0087] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0088] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, device, product or end including a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units inherent to these processes, methods, products or ends.

[0089] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0090] The present invention discloses a vehicle-side roadside perception method, system and device based on monocular vision. Before lane perception, by analyzing the first scene data and judging the set roadside perception conditions, perception in inappropriate scenes is avoided, which is conducive to improving the accuracy and reliability of subsequent operations; then, by combining the second scene data collected in real time with the basic road information, not only the real-time monitoring of the relationship between the vehicle and the road boundary during driving is realized, but also the real-time and accuracy of the perception results are improved; finally, by accurately calculating the roadside ratio information, the position of the vehicle on the road can be monitored in real time, and the driver can be provided with intuitive vehicle position feedback, which helps the driver to correct the vehicle's driving direction in time, avoid deviation from the road or collision accidents, thereby significantly improving driving safety and road driving stability. The following are detailed descriptions.

[0091] Embodiment 1

[0092] See also Figure 1 , Figure 1 : is a flow chart of a vehicle-side roadside perception method based on monocular vision disclosed in an embodiment of the present invention. Figure 1 The vehicle-side roadside perception method based on monocular vision described above can be applied to a vehicle-side roadside perception device or system based on monocular vision, and the embodiments of the present invention do not limit this. Figure 1 As shown, the vehicle-side roadside perception method based on monocular vision may include the following operations:

[0093] 101. Collect first scene data of a target vehicle in a current scene before it is started, where the first scene data includes first vehicle front data within a preset viewing angle range in front of the target vehicle and first vehicle posture information of the target vehicle.

[0094] In the embodiment of the present invention, the collection of the first scene data provides necessary initial data for the subsequent perception process.

[0095] In an embodiment of the present invention, the first scene data may be collected by a monocular camera device installed on the target vehicle, such as a monocular camera; the first vehicle posture information may be collected by a MEMS system installed on the target vehicle as a vehicle posture sensor.

[0096] 102. After determining that the first scene data indicates that a preset roadside perception condition is currently satisfied, the first scene data is analyzed to obtain basic road information of the front scene.

[0097] In the embodiment of the present invention, the basic road information at least includes the road width and the target camera angle corresponding to the monocular camera device.

[0098] In an embodiment of the present invention, by analyzing the first scene data, it is possible to determine whether the current scene meets the preset roadside perception conditions. The analysis of the first scene data and the setting and judgment of the roadside perception conditions avoid perception in inappropriate scenarios, which is conducive to improving the accuracy and reliability of subsequent operations.

[0099] 103. Collect in real time second scene data of the target vehicle in the current scene after the target vehicle starts to drive, where the second scene data includes second vehicle front data and second vehicle posture information.

[0100] In the embodiment of the present invention, the first and second data in front of the vehicle and the second data in front of the vehicle, and the first vehicle posture information and the second vehicle posture information are used to refer to different data collection times.

[0101] In the embodiment of the present invention, it should be noted that the second scene data refers to the data collected in real time after the target vehicle is started, which is relatively dynamic time-sensitive data. The time-sensitive data is used to refer to the second scene data collected in real time according to the driving of the target vehicle, and the second scene data is continuous and changing data, so as to realize continuous data collection or detection for the current scene; correspondingly, the first scene data collected in the pre-step 101 is "relatively static" recorded data. After the first scene data is collected, the first scene data will no longer change or update.

[0102] 104. According to a preset roadside perception algorithm, in combination with basic road information and second scene data, target measurement information is calculated, where the target measurement information includes roadside ratio information of the target vehicle and the road.

[0103] In an embodiment of the present invention, the roadside ratio information is used to indicate the deviation attribute of the target vehicle on the road, and the deviation attribute includes any one of a left-biased attribute indicating that the target vehicle is located on the left side of the road, a right-biased attribute indicating that the target vehicle is located on the right side of the road, and a centered attribute indicating that the target vehicle is located in the middle of the road.

[0104] In the embodiment of the present invention, it should be noted that the vehicle-side roadside perception method based on monocular vision can be applied to the corresponding vehicle-side roadside perception system based on monocular vision. Correspondingly, in actual use, the system can be installed on the customer's vehicle side. Specifically, at the vehicle side, in order to implement the above steps 101 to 104, the vehicle side is at least configured with a sensor device, a core processing unit and a power supply module.

[0105] In an embodiment of the present invention, the sensing device may include a core sensor composed of a monocular camera and a vehicle posture sensor composed of a MEMS system; wherein:

[0106] The monocular camera is installed on the vehicle side as a core sensor; the monocular camera is used to observe the scene in front of the vehicle and the road signs on both sides of the road, and transmit the obtained video data to the core processing unit;

[0107] The MEMS system is installed on the vehicle side as a vehicle posture sensor; the MEMS system is used to obtain vehicle posture information during vehicle movement and output the obtained data to the core processing unit;

[0108] The power supply module is used to provide working power for the sensor equipment (including the above-mentioned monocular camera, MEMS system) and the core processing unit.

[0109] It can be seen that implementation Figure 1 The described vehicle-side roadside perception method based on monocular vision, before lane perception, avoids perception in inappropriate scenarios by analyzing the first scene data and judging the set roadside perception conditions, which is beneficial to improving the accuracy and reliability of subsequent operations; then, by combining the second scene data collected in real time with the basic road information, it not only realizes the real-time monitoring of the relationship between the vehicle and the road boundary during driving, but also helps to improve the real-time and accuracy of the perception results; finally, by accurately calculating the roadside ratio information, it can monitor the position of the vehicle on the road in real time and provide the driver with intuitive vehicle position feedback, which helps the driver to correct the vehicle's driving direction in time to avoid deviation from the road or collision accidents, thereby significantly improving driving safety and road driving stability.

[0110] Embodiment 2

[0111] See also Figure 2 , Figure 2 is a flow chart of another vehicle-side roadside perception method based on monocular vision disclosed in an embodiment of the present invention. Figure 2 The vehicle-side roadside perception method based on monocular vision described above can be applied to a vehicle-side roadside perception system or device based on monocular vision, and the embodiments of the present invention do not limit this. Figure 2 As shown, the vehicle-side roadside perception method based on monocular vision may include the following operations:

[0112] 201. Collect first scene data of a target vehicle in a current scene before it is started, where the first scene data includes first vehicle front data within a preset viewing angle range in front of the target vehicle and first vehicle posture information of the target vehicle.

[0113] In an embodiment of the present invention, two roadside signs are set on both sides of the road in the current scene, one roadside sign is set on each side of the road, and the two roadside signs are set on the same horizontal line, which is perpendicular to the road; the preset viewing angle range is the device photography range of the monocular camera device.

[0114] In the embodiment of the present invention, the two roadside signs are provided as important reference objects for determining the relationship between the vehicle position and the road boundary.

[0115] 202. Determine whether the first vehicle front data indicates that the two roadside signs can be identified within the photographing range of the device.

[0116] In the embodiment of the present invention, when the judgment result of step 202 is yes, step 203 is triggered, and then step 205 is triggered; when the judgment result of step 202 is no, step 204 is triggered, and then step 205 is triggered.

[0117] 203. Determine whether the first scene data indicates that a preset roadside perception condition is currently satisfied.

[0118] 204. Perform posture adjustment on the target vehicle; and after determining that the monocular camera device is adjusted to be able to simultaneously identify the two roadside signs, update the first scene data, and determine that the first scene data indicates that a preset roadside perception condition is currently met.

[0119] 205. After determining that the first scene data indicates that a preset roadside perception condition is currently satisfied, the first scene data is analyzed to obtain basic road information of the front scene.

[0120] 206. Collect in real time second scene data of the target vehicle in the current scene after the target vehicle starts driving, the second scene data including second vehicle front data and second vehicle posture information.

[0121] 207. According to a preset roadside perception algorithm, in combination with basic road information and second scene data, target measurement information is calculated, where the target measurement information includes roadside ratio information of the target vehicle and the road.

[0122] In the embodiment of the present invention, for other descriptions of step 201 and step 205 to step 207, please refer to other specific descriptions of step 101 and step 102 to step 104 in embodiment 1, and the embodiment of the present invention will not be repeated here.

[0123] It can be seen that implementation Figure 2 The described vehicle-side roadside perception method based on monocular vision realizes accurate judgment and adjustment of roadside perception conditions by introducing roadside signs and specific judgment logic. Among them, a processing mechanism is also set up for failure to meet roadside perception conditions, which improves the reliability and accuracy of subsequent roadside perception by intelligently adjusting the vehicle posture to meet the perception conditions.

[0124] In an optional embodiment, the manner in which the above step 204 performs posture adjustment on the target vehicle specifically includes:

[0125] Determine, based on the first vehicle posture information, the current position of the target vehicle on the road, the offset angle relative to the road, and the posture parameters corresponding to the vehicle-roadside sign distance;

[0126] Determine the maximum recognition distance of the monocular camera device according to the device photographing range of the monocular camera device;

[0127] Calculating the difference between the maximum recognition distance and the vehicle-roadside sign distance, and taking the absolute value of the difference to obtain a first adjustment distance;

[0128] According to the first adjustment distance, in combination with the current position and the offset angle, determine the second adjustment distance and the corresponding distance adjustment direction, the body adjustment angle and the corresponding angle adjustment direction for controlling the target vehicle to move; the distance adjustment direction includes a first direction close to the roadside sign or a second direction away from the roadside sign; the angle adjustment direction includes a clockwise direction or a counterclockwise direction;

[0129] According to the second adjustment distance and its corresponding distance adjustment direction, as well as the vehicle body adjustment angle and its corresponding angle adjustment direction, the target vehicle is controlled to perform posture adjustment, and when it is determined that the first vehicle-front data indicates that the two roadside signs can be identified within the photographic range of the equipment, it is determined that the posture adjustment operation for the target vehicle is completed.

[0130] In this optional embodiment, specifically, it is assumed that the specific situation of the photographic range of the monocular camera device is a fan-shaped range with a radius of 15 meters and an angle of 100°; at this time, if it is set that there is no need to adjust the angle of the target vehicle body, and the target vehicle is parked at a distance of 20 meters from the roadside sign, correspondingly, the posture adjustment operation performed on the target vehicle is specifically to control the target vehicle to approach the roadside sign by at least 5 meters.

[0131] In this optional embodiment, optionally, if it is assumed that the target vehicle is parked within 10 meters of the roadside sign, there is no need to control the displacement of the target vehicle at this time, and the angle of the target vehicle relative to the road side line is 60°, and the target vehicle is parked on the right side line of the road. At this time, the monocular camera device can only recognize the roadside sign arranged on the right side line of the road, and it is necessary to control the target vehicle to rotate left / counterclockwise at least 40° to simultaneously recognize two roadside signs. In addition, for the situation where the target vehicle needs to be controlled to move forward and rotate the body at the same time, it is a combination of the above two situations, which will not be repeated here; and for the above-mentioned related parameters, such as the equipment photography range, the specific adjustment angle value of the body angle adjustment, the specific driving distance of the body, the driving direction (close to the roadside sign or away from the roadside sign), etc., it needs to be adjusted according to the actual situation, and the embodiment of the present invention is not limited.

[0132] It can be seen that in this optional embodiment, for the situation where the target vehicle needs to be adjusted in posture, the current position of the target vehicle on the road, the offset angle relative to the road and the posture parameters corresponding to the vehicle-roadside sign distance can be intelligently determined based on the first vehicle posture information as the basic information, and then the second adjustment distance and its corresponding distance adjustment direction, the body adjustment angle and its corresponding angle adjustment direction and other multi-dimensional information that ultimately control the movement of the target vehicle can be accurately determined, thereby improving the accuracy of parameter determination for performing posture adjustment on the target vehicle and the accuracy and reliability of performing posture adjustment.

[0133] In another optional embodiment, the first vehicle posture information includes a device position corresponding to a monocular camera device;

[0134] The above step 102 analyzes the first scene data to obtain basic road information of the front scene in the following manner:

[0135] According to a preset target recognition algorithm, the road side line is used as the recognition object, and target recognition is performed on the first vehicle front data to obtain the road width corresponding to the road in the current scene;

[0136] Read the photographic range of the monocular camera device in the current scene and its corresponding photographic angle;

[0137] Taking the device position as the reference point and the road sideline as the reference direction, perform angle division on the photography angle to obtain the angle between the monocular camera device and the road sideline on the left, which is used as the target photography angle;

[0138] The road width and the target camera angle are determined as the basic road information of the front scene.

[0139] It can be seen that in this optional embodiment, through the preset target recognition algorithm, the road side line is taken as the recognition object, and target recognition is performed on the first vehicle front data, so as to accurately obtain the road width corresponding to the road in the current scene. This step provides the system with key road size information, which is helpful for subsequent road information analysis and processing efficiency; in addition, the relative position relationship between the camera device and the road side line, as well as the shooting angle of the camera device, can also be considered to ensure the accuracy and scientificity of the target camera angle; finally, by refining the first vehicle posture information and accurately obtaining key information such as road width and target camera angle, the accuracy of determining basic road information is improved.

[0140] In another optional embodiment, the method of calculating the target measurement information according to the preset roadside perception algorithm in combination with the basic road information and the second scene data specifically includes:

[0141] The two roadside signs are divided into a first sign located on the left side of the road and a second sign located on the right side of the road; wherein, after the target vehicle starts to travel, the roadside sign that the target vehicle travels away from is the first sign, and the roadside sign that the target vehicle travels closer to is the second sign;

[0142] Determine, based on the second vehicle front data, the time difference between the two roadside signs disappearing one after another after the target vehicle starts to move;

[0143] determining an average driving speed of the target vehicle according to the second vehicle posture information;

[0144] According to the time difference, average driving speed, road width and the preset roadside ratio calculation formula, the target roadside ratio value is determined as the target measurement information.

[0145] It can be seen that in this optional embodiment, by introducing specific roadside sign division, time difference measurement, average driving speed calculation and roadside ratio calculation formula, accurate measurement of the relationship between the target vehicle and the road boundary during driving is achieved, which is beneficial to improving the calculation accuracy and reliability of roadside perception.

[0146] In another optional embodiment, the method further comprises:

[0147] Determine the disappearance order of the two roadside signs according to the target roadside ratio and the second vehicle front data;

[0148] Determine the deviation information of the target vehicle on the road according to the disappearance order of the two roadside signs;

[0149] When the deviation information indicates that the target vehicle is located in the middle of the road, a safety warning is generated for the target vehicle to prompt the driver of the target vehicle to make driving adjustments;

[0150] Among them, when the second data in front of the vehicle indicates that the first mark disappears first, the deviation information indicates that the target vehicle is located on the right side of the road; when the second data in front of the vehicle indicates that the second mark disappears first, the deviation information indicates that the target vehicle is located on the left side of the road; when the second data in front of the vehicle indicates that the first mark and the second mark disappear at the same time, the deviation information indicates that the target vehicle is located in the middle of the road.

[0151] It can be seen that in this optional embodiment, the deviation information of the target measurement on the road can be accurately obtained according to the order of disappearance of the roadside signs. The deviation information directly reflects the position of the target vehicle on the road, providing the driver with intuitive driving assistance information; further, when the deviation information indicates that the target vehicle is located in the middle of the road, a safety warning can be generated for the target vehicle. The setting of the warning mechanism can prevent the vehicle from deviating from the road or causing a collision accident, and promptly remind the driver to make driving adjustments, thereby helping to improve the user's driving safety.

[0152] In yet another optional embodiment, the determination method corresponding to the roadside ratio calculation formula includes:

[0153] Determine the first node data and the second node data according to the second vehicle front data and the second vehicle posture information; the first node data includes the first recording time and the corresponding first vehicle position; the second node data includes the second recording time and the corresponding second vehicle position; the first recording time is the time corresponding to the initial disappearance of the first mark after the target vehicle starts to drive; the second recording time is the time corresponding to the initial disappearance of the second mark after the target vehicle starts to drive;

[0154] Establish a coordinate system with the first vehicle position as the origin, the direction corresponding to the road sideline as the longitudinal axis, and the direction perpendicular to the road sideline as the transverse axis; and determine the first coordinate, the second coordinate, and the third coordinate of the first identifier, the second identifier, and the second vehicle position in the coordinate system;

[0155] Determine the driving deviation angle of the target vehicle according to the third coordinate, where the driving deviation angle is the angle between the longitudinal axis and the driving distance; the driving distance is the straight-line distance between the first vehicle position and the second vehicle position;

[0156] The roadside ratio calculation formula is determined according to the first coordinate, the second coordinate, the third coordinate, the driving distance, the driving deviation angle and the set basic roadside ratio formula.

[0157] In this alternative embodiment, see Figure 6 , Figure 6 is a working schematic diagram of a vehicle-side roadside perception system based on monocular vision disclosed in an embodiment of the present invention; Figure 6 As shown, Figure 6The gray square where point O is located represents the perception system, which is composed of a monocular camera and a MEMS processing system. The gray square can also refer to the target vehicle, which is simplified here by a square. The gray square with diagonal lines represents the position reached by the target vehicle after a certain period of movement, represented by the letter O'. The gray origins L and R on both sides are the road signs on the left side of the road (the first mark mentioned above) and the road signs on the right side (the second mark mentioned above), respectively. The gray lines on both sides represent the side lines of the road. With point O as the origin and the forward direction as the y-axis, a carrier coordinate system is established, then O = (0, 0), L = (x L ,y L ), R=(x R ,y R ). The field of view of the monocular camera is θ; the road width is S. →V represents the motion vector of the target vehicle, which includes the motion direction ε and speed v of the target vehicle. The data is provided by the MEMS processing system.

[0158] In this optional embodiment, the target vehicle moves at a speed v in the direction of the arrow. At a certain moment, the left road sign L disappears from the field of vision (here, it refers to the node where the road sign L is about to disappear). At this time, the vehicle is at point O. A straight line perpendicular to the road direction is drawn through O. The straight line intersects the edges of the two sides of the road at points D and A. Assume that the ratio of the target vehicle to the right side of the road to the left side of the road is λ, then After the vehicle continues to move for t time, it is observed that the right road sign R disappears from the field of vision. At this time, the vehicle is located at point O'=(x1, y1).

[0159] Combination Figure 6 , the following relationship exists:

[0160]

[0161] O′=O+vt (2)

[0162]

[0163] Will Substitute into the above equations (1)-(3), and solve the equations simultaneously, eliminating (x L ,y L ,x R ,y R ), the road side ratio calculation formula can be obtained:

[0164]

[0165] In this optional embodiment, in the roadside ratio calculation formula, corresponding to the above related description, λ corresponds to the ratio between the target vehicle to the right side of the road and the target vehicle to the left side of the road; is the motion vector of the target vehicle, which includes the driving deviation angle ε and the average driving speed v of the target vehicle; t is the moving time corresponding to the average driving speed; S is the road width, and θ is the target camera angle.

[0166] It can be seen that in this optional embodiment, by observing the disappearance time difference of the road signs on both sides of the road, combined with the target vehicle's own movement, the proportional relationship between the target vehicle and the roadside is obtained, providing a basis for safe driving of the vehicle. At the same time, this method does not rely on depth data, and gets rid of the dependence of visual perception on binocular cameras; secondly, this method does not recognize the geometric features of the target, which reduces the amount of visual perception calculations and improves the perception speed; finally, the perception system is installed on the vehicle side and does not need to rely on the collaborative calculation of external equipment, which improves the autonomy of the system and reduces the cost of the system.

[0167] It can be seen that in this optional embodiment, the roadside ratio calculation formula that is set comprehensively considers key information such as vehicle position, roadside sign position, and also considers the impact of driving deviation angle on the roadside ratio, thereby ensuring the accuracy and reliability of the calculation results.

[0168] Embodiment 3

[0169] See also Figure 3 , Figure 3 It is a structural diagram of a vehicle-side roadside perception system based on monocular vision disclosed in an embodiment of the present invention. The vehicle-side roadside perception system based on monocular vision can be a vehicle-side roadside perception terminal, device, apparatus or server based on monocular vision. The server can be a local server, a remote server, or a cloud server (also known as a cloud server). When the server is a non-cloud server, the non-cloud server can communicate with the cloud server, which is not limited in the embodiment of the present invention. Figure 3 As shown, the vehicle-side roadside perception system based on monocular vision may include a data acquisition module 301, a data analysis module 302 and a calculation module 303, wherein:

[0170] The data acquisition module 301 is used to collect the first scene data of the target vehicle in the current scene before it starts, and the first scene data includes the first vehicle front data within a preset viewing angle range in front of the target vehicle and the first vehicle posture information of the target vehicle.

[0171] The data analysis module 302 is used to analyze the first scene data after determining that the first scene data indicates that the preset roadside perception conditions are currently met, and obtain basic road information of the front scene. The basic road information at least includes the road width and the target camera angle corresponding to the monocular camera device.

[0172] The data acquisition module 301 is also used to collect in real time the second scene data of the current scene after the target vehicle starts driving, and the second scene data includes the second vehicle front data and the second vehicle posture information.

[0173] The calculation module 303 is used to calculate the target measurement information according to the preset roadside perception algorithm, combined with the basic road information and the second scene data, wherein the target measurement information includes the roadside ratio information of the target vehicle and the road; wherein the roadside ratio information is used to indicate the deviation attribute of the target vehicle on the road, and the deviation attribute includes any one of the left deviation attribute indicating that the target vehicle is located on the left side of the road, the right deviation attribute indicating that the target vehicle is located on the right side of the road, and the centering attribute indicating that the target vehicle is located in the middle of the road.

[0174] It can be seen that implementation Figure 3 The described vehicle-side roadside perception system based on monocular vision, before lane perception, avoids perception in inappropriate scenarios by analyzing the first scene data and judging the set roadside perception conditions, which is beneficial to improving the accuracy and reliability of subsequent operations; then, by combining the second scene data collected in real time with the basic road information, it not only realizes the real-time monitoring of the relationship between the vehicle and the road boundary during driving, but also helps to improve the real-time and accuracy of the perception results; finally, by accurately calculating the roadside ratio information, it can monitor the position of the vehicle on the road in real time and provide the driver with intuitive vehicle position feedback, which helps the driver to correct the vehicle's driving direction in time to avoid deviation from the road or collision accidents, thereby significantly improving driving safety and road driving stability.

[0175] In an optional embodiment, two roadside signs are set on both sides of the current scene, one roadside sign is set on each side of the road, and the two roadside signs are set on the same horizontal line, and the horizontal line is perpendicular to the road; the preset viewing angle range is the device photography range of the monocular camera device;

[0176] See also Figure 4 , Figure 4 FIG. 1 is a schematic diagram of another vehicle-side roadside perception system based on monocular vision disclosed in an embodiment of the present invention. Figure 4 As shown, the system also includes a judgment module 304, a determination module 305, a posture adjustment module 306 and a data update module 307:

[0177] The judgment module 304 is used to judge whether the first vehicle front data indicates that the two roadside signs can be identified within the photographing range of the device after the data collection module 301 collects the first scene data of the current scene before the target vehicle starts.

[0178] The determination module 305 is used to determine that the first scene data currently satisfies a preset roadside perception condition when the determination module 304 determines that the first vehicle front data indicates that the two roadside signs can be identified within the photographing range of the device.

[0179] The posture adjustment module 306 is used to perform posture adjustment on the target vehicle when the judgment module 304 judges that the first vehicle front data indicates that the two roadside signs cannot be identified within the device photography range.

[0180] The data updating module 307 is used to update the first scene data after determining that the monocular camera device is adjusted to be able to simultaneously identify the two roadside signs, and to determine that the first scene data indicates that a preset roadside perception condition is currently met.

[0181] It can be seen that in this optional embodiment, accurate judgment and adjustment of roadside perception conditions are achieved by introducing roadside signs and specific judgment logic; wherein, a processing mechanism is also provided for failure to meet roadside perception conditions, which improves the reliability and accuracy of subsequent roadside perception by intelligently adjusting the vehicle posture to meet the perception conditions.

[0182] In another optional embodiment, the first vehicle posture information includes a device position corresponding to a monocular camera device;

[0183] The data analysis module 302 analyzes the first scene data to obtain basic road information of the front scene in the following manner:

[0184] According to a preset target recognition algorithm, the road side line is used as the recognition object, and target recognition is performed on the first vehicle front data to obtain the road width corresponding to the road in the current scene;

[0185] Read the photographic range of the monocular camera device in the current scene and its corresponding photographic angle;

[0186] Taking the device position as the reference point and the road sideline as the reference direction, perform angle division on the photography angle to obtain the angle between the monocular camera device and the road sideline on the left, which is used as the target photography angle;

[0187] The road width and the target camera angle are determined as the basic road information of the front scene.

[0188] It can be seen that in this optional embodiment, through the preset target recognition algorithm, the road side line is taken as the recognition object, and target recognition is performed on the first vehicle front data, so as to accurately obtain the road width corresponding to the road in the current scene. This step provides the system with key road size information, which is helpful for subsequent road information analysis and processing efficiency; in addition, the relative position relationship between the camera device and the road side line, as well as the shooting angle of the camera device, can also be considered to ensure the accuracy and scientificity of the target camera angle; finally, by refining the first vehicle posture information and accurately obtaining key information such as road width and target camera angle, the accuracy of determining basic road information is improved.

[0189] In another optional embodiment, the calculation module 303 calculates the target measurement information according to the preset roadside perception algorithm in combination with the basic road information and the second scene data, specifically including:

[0190] The two roadside signs are divided into a first sign located on the left side of the road and a second sign located on the right side of the road; wherein, after the target vehicle starts to travel, the roadside sign that the target vehicle travels away from is the first sign, and the roadside sign that the target vehicle travels closer to is the second sign;

[0191] Determine, based on the second vehicle front data, the time difference between the two roadside signs disappearing one after another after the target vehicle starts to move;

[0192] determining an average driving speed of the target vehicle according to the second vehicle posture information;

[0193] According to the time difference, average driving speed, road width and the preset roadside ratio calculation formula, the target roadside ratio value is determined as the target measurement information.

[0194] It can be seen that in this optional embodiment, by introducing specific roadside sign division, time difference measurement, average driving speed calculation and roadside ratio calculation formula, accurate measurement of the relationship between the target vehicle and the road boundary during driving is achieved, which is beneficial to improving the calculation accuracy and reliability of roadside perception.

[0195] In another optional embodiment, the determination module 305 is further used to determine the disappearance order of the two roadside signs according to the target roadside ratio and the second vehicle front data;

[0196] The determination module 305 is further used to determine the deviation information of the target vehicle on the road according to the disappearance order of the two roadside signs;

[0197] like Figure 4 As shown, the system also includes a safety warning module 308, wherein:

[0198] A safety warning module 308, for generating a safety warning for the target vehicle when the deviation information indicates that the target vehicle is located in the middle of the road, so as to prompt the driver of the target vehicle to make driving adjustments;

[0199] Among them, when the second data in front of the vehicle indicates that the first mark disappears first, the deviation information indicates that the target vehicle is located on the right side of the road; when the second data in front of the vehicle indicates that the second mark disappears first, the deviation information indicates that the target vehicle is located on the left side of the road; when the second data in front of the vehicle indicates that the first mark and the second mark disappear at the same time, the deviation information indicates that the target vehicle is located in the middle of the road.

[0200] It can be seen that in this optional embodiment, the deviation information of the target measurement on the road can be accurately obtained according to the order of disappearance of the roadside signs. The deviation information directly reflects the position of the target vehicle on the road, providing the driver with intuitive driving assistance information; further, when the deviation information indicates that the target vehicle is located in the middle of the road, a safety warning can be generated for the target vehicle. The setting of the warning mechanism can prevent the vehicle from deviating from the road or causing a collision accident, and promptly remind the driver to make driving adjustments, thereby helping to improve the user's driving safety.

[0201] In yet another optional embodiment, the determination method corresponding to the roadside ratio calculation formula specifically includes:

[0202] Determine the first node data and the second node data according to the second vehicle front data and the second vehicle posture information; the first node data includes the first recording time and the corresponding first vehicle position; the second node data includes the second recording time and the corresponding second vehicle position; the first recording time is the time corresponding to the initial disappearance of the first mark after the target vehicle starts to drive; the second recording time is the time corresponding to the initial disappearance of the second mark after the target vehicle starts to drive;

[0203] Establish a coordinate system with the first vehicle position as the origin, the direction corresponding to the road sideline as the longitudinal axis, and the direction perpendicular to the road sideline as the transverse axis; and determine the first coordinate, the second coordinate, and the third coordinate of the first identifier, the second identifier, and the second vehicle position in the coordinate system;

[0204] Determine the driving deviation angle of the target vehicle according to the third coordinate, where the driving deviation angle is the angle between the longitudinal axis and the driving distance; the driving distance is the straight-line distance between the first vehicle position and the second vehicle position;

[0205] The roadside ratio calculation formula is determined according to the first coordinate, the second coordinate, the third coordinate, the driving distance, the driving deviation angle and the set basic roadside ratio formula.

[0206] In this optional embodiment, the formula corresponding to the roadside ratio calculation formula is specifically:

[0207]

[0208] Among them, λ corresponds to the ratio between the target vehicle to the right side of the road and the target vehicle to the left side of the road; is the motion vector of the target vehicle, which includes the driving deviation angle ε and the average driving speed v of the target vehicle; t is the moving time corresponding to the average driving speed; S is the road width, and θ is the target camera angle.

[0209] It can be seen that in this optional embodiment, the roadside ratio calculation formula that is set comprehensively considers key information such as vehicle position, roadside sign position, and also considers the impact of driving deviation angle on the roadside ratio, thereby ensuring the accuracy and reliability of the calculation results.

[0210] Embodiment 4

[0211] See also Figure 5 , Figure 5 Schematic diagram of the structure of a vehicle-side roadside perception device based on monocular vision disclosed in an embodiment of the present invention. Figure 5 As shown, the vehicle-side roadside perception device based on monocular vision may include:

[0212] A memory 401 storing executable program codes;

[0213] a processor 402 coupled to the memory 401;

[0214] The processor 402 calls the executable program code stored in the memory 401 to execute the steps in the vehicle-side roadside perception method based on monocular vision described in the first embodiment of the present invention or the second embodiment of the present invention.

[0215] Embodiment 5

[0216] An embodiment of the present invention discloses a computer storage medium, which stores computer instructions. When the computer instructions are called, they are used to execute the steps in the vehicle-side roadside perception method based on monocular vision described in Embodiment 1 or Embodiment 2 of the present invention.

[0217] Embodiment 6

[0218] An embodiment of the present invention discloses a computer program product, which includes a non-transitory computer storage medium storing a computer program, and the computer program is operable to enable a computer to execute the steps in the monocular vision-based vehicle-side roadside perception method described in Example 1 or Example 2.

[0219] The device embodiments described above are only illustrative, wherein the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, i.e., they may be located in one place, or they may be distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative work.

[0220] Through the specific description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the above technical solution can be essentially or partly contributed to the prior art in the form of a software product, and the computer software product can be stored in a computer storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0221] Finally, it should be noted that the monocular vision-based vehicle-side roadside perception method, system, and device disclosed in the embodiments of the present invention are only preferred embodiments of the present invention, which are only used to illustrate the technical solution of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, it should be understood by those skilled in the art that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features therein can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A vehicle-side roadside perception method based on monocular vision, characterized in that: The method comprises: Collecting first scene data of the target vehicle in the current scene before it starts, the first scene data including first vehicle front data within a preset viewing angle in front of the target vehicle and first vehicle posture information of the target vehicle; After determining that the first scene data indicates that a preset roadside perception condition is currently satisfied, analyzing the first scene data to obtain basic road information of the front scene, the basic road information at least including a road width and a target camera angle corresponding to a monocular camera device; Real-time collection of second scene data of the current scene after the target vehicle starts to travel, the second scene data including second vehicle front data and second vehicle posture information; According to a preset roadside perception algorithm, in combination with the basic road information and the second scene data, target measurement information is calculated, the target measurement information including roadside ratio information of the target vehicle and the road; wherein the roadside ratio information is used to indicate a deviation attribute of the target vehicle on the road, the deviation attribute including any one of a left deviation attribute indicating that the target vehicle is located on the left side of the road, a right deviation attribute indicating that the target vehicle is located on the right side of the road, and a centering attribute indicating that the target vehicle is located in the middle of the road.

2. The vehicle-side roadside perception method based on monocular vision according to claim 1 is characterized in that: There are two roadside signs on both sides of the current scene, one roadside sign is set on each side of the road, and the two roadside signs are set on the same horizontal line, and the horizontal line is perpendicular to the road; the preset viewing angle range is the device photography range of the monocular camera device; After collecting the first scene data of the current scene before the target vehicle is started, the method further includes: Determining whether the first vehicle-front data indicates that the two roadside signs can be identified within the photographic range of the device, and when it is determined that the first vehicle-front data indicates that the two roadside signs can be identified within the photographic range of the device, determining that the first scene data indicates that a preset roadside perception condition is currently satisfied; When it is determined that the first vehicle-front data indicates that the two roadside signs cannot be identified within the photographic range of the device, the position adjustment is performed on the target vehicle; and after determining that the monocular camera device is adjusted to be able to simultaneously identify the two roadside signs, the first scene data is updated, and it is determined that the first scene data indicates that the preset roadside perception conditions are currently met.

3. The vehicle-side roadside perception method based on monocular vision according to claim 1 or 2, characterized in that: The first vehicle posture information includes a device position corresponding to a monocular camera device; The analyzing the first scene data to obtain basic road information of the front scene includes: According to a preset target recognition algorithm, taking the road side line as the recognition object, performing target recognition on the first vehicle front data to obtain the road width corresponding to the road in the current scene; Reading the photographing range of the monocular camera device in the current scene and its corresponding photographing angle; Taking the device position as a reference point and taking the direction parallel to the road side line as a reference direction, performing angle division on the photography angle, and obtaining the angle between the monocular camera device and the road side line on the left as the target photography angle; The road width and the target camera angle are determined as basic road information of the front scene.

4. The vehicle-side roadside perception method based on monocular vision according to claim 2 is characterized in that: The calculating target measurement information according to a preset roadside perception algorithm in combination with the basic road information and the second scene data includes: The two roadside signs are divided into a first sign located on the left side of the road and a second sign located on the right side of the road; wherein, after the target vehicle starts to travel, the roadside sign that the target vehicle travels away from is the first sign, and the roadside sign that the target vehicle travels towards is the second sign; Determine, based on the second vehicle front data, a time difference between the two roadside signs disappearing in sequence after the target vehicle starts to travel; determining an average travel speed of the target vehicle according to the second vehicle posture information; According to the time difference, the average driving speed, the road width, and a preset roadside ratio calculation formula, a target roadside ratio value is determined as target measurement information.

5. The vehicle-side roadside perception method based on monocular vision according to claim 4 is characterized in that: The method further comprises: Determining a disappearance order of the two roadside signs according to the target roadside ratio and the second vehicle front data; Determining the deviation information of the target vehicle on the road according to the disappearance order of the two roadside signs; When the deviation information indicates that the target vehicle is located in the middle of the road, a safety warning is generated for the target vehicle to prompt the driver of the target vehicle to make driving adjustments; Among them, when the second data in front of the vehicle indicates that the first mark disappears first, the deviation information indicates that the target vehicle is located on the right side of the road; when the second data in front of the vehicle indicates that the second mark disappears first, the deviation information indicates that the target vehicle is located on the left side of the road; when the second data in front of the vehicle indicates that the first mark and the second mark disappear at the same time, the deviation information indicates that the target vehicle is located in the middle of the road.

6. The vehicle-side roadside perception method based on monocular vision according to claim 4 or 5, characterized in that: The determination method corresponding to the roadside ratio calculation formula includes: Determine the first node data and the second node data according to the second vehicle front data and the second vehicle posture information; the first node data includes a first recording time and a first vehicle position corresponding thereto; the second node data includes a second recording time and a second vehicle position corresponding thereto; the first recording time is a time corresponding to the initial disappearance of the first mark after the target vehicle starts to travel; the second recording time is a time corresponding to the initial disappearance of the second mark after the target vehicle starts to travel; Establish a coordinate system with the first vehicle position as the origin, the direction corresponding to the road sideline as the longitudinal axis, and the direction perpendicular to the road sideline as the transverse axis; and determine the first coordinate, the second coordinate, and the third coordinate of the first identifier, the second identifier, and the second vehicle position in the coordinate system; Determine the driving deviation angle of the target vehicle according to the third coordinate, wherein the driving deviation angle is the included angle between the longitudinal axis and the driving distance; the driving distance is the straight-line distance between the first vehicle position and the second vehicle position; A roadside ratio calculation formula is determined according to the first coordinate, the second coordinate, the third coordinate, the driving distance, the driving deviation angle, and a set basic roadside ratio formula.

7. The vehicle-side roadside perception method based on monocular vision according to claim 6 is characterized in that: The formula corresponding to the roadside ratio calculation formula is specifically: Wherein, λ corresponds to the ratio between the target vehicle to the right side of the road and the target vehicle to the left side of the road; is the motion vector of the target vehicle, which includes the driving deviation angle ε and the average driving speed v of the target vehicle; t is the moving time corresponding to the average driving speed; S is the road width of the road, and θ is the target camera angle.

8. A vehicle-side roadside perception system based on monocular vision, the system comprising: A data acquisition module, used to collect first scene data of the target vehicle in the current scene before it is started, wherein the first scene data includes first vehicle front data within a preset viewing angle range in front of the target vehicle and first vehicle posture information of the target vehicle; A data analysis module, configured to analyze the first scene data to obtain basic road information of the front scene after determining that the first scene data indicates that a preset roadside perception condition is currently satisfied, wherein the basic road information includes at least a road width and a target camera angle corresponding to a monocular camera device; The data acquisition module is further used to collect in real time the second scene data of the current scene after the target vehicle starts to travel, wherein the second scene data includes the second vehicle front data and the second vehicle posture information; A calculation module is used to calculate target measurement information according to a preset roadside perception algorithm, in combination with the basic road information and the second scene data, wherein the target measurement information includes roadside ratio information of the target vehicle and the road; wherein the roadside ratio information is used to indicate a deviation attribute of the target vehicle on the road, and the deviation attribute includes any one of a left deviation attribute indicating that the target vehicle is located on the left side of the road, a right deviation attribute indicating that the target vehicle is located on the right side of the road, and a centering attribute indicating that the target vehicle is located in the middle of the road.

9. A vehicle-side roadside perception device based on monocular vision, characterized in that: The device comprises: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the vehicle-side roadside perception method based on monocular vision as described in any one of claims 1-7.

10. A computer storage medium, characterized in that: The computer storage medium stores computer instructions, which, when called, are used to execute the vehicle-side roadside perception method based on monocular vision as described in any one of claims 1 to 7.

Images