CN116386000A - Method and system for measuring obstacle distance based on high-precision map and monocular camera - Google Patents

Abstract

The present invention provides a method and system for measuring obstacle distance based on a high-precision map and a monocular camera. The method includes: identifying elements on the high-precision map; The obstacle in the captured image is matched with the elements on the recognized high-precision map; it is judged whether the height information of the obstacle is known; if the height information of the obstacle is known, it is based on the principle of monocular distance measurement and the obstacle height information to obtain the distance between the obstacle and the vehicle. The invention can solve the technical problem in the prior art that the distance of the obstacle cannot be acquired solely by a monocular camera, the size and direction of the obstacle must be known, or the distance of the obstacle can only be obtained by using a binocular camera.

Description

Method and system for measuring obstacle distance based on high-precision map and monocular camera

technical field

The present invention relates to the technical field of intelligent driving, in particular to a method and system for measuring obstacle distances based on a high-precision map and a monocular camera.

Background technique

Autonomous driving technology is the product of the development and integration of automotive electronics, intelligent control, and the Internet. Its principle is that the autonomous driving system obtains the vehicle's own information and surrounding environment information through the perception system, and the collected data information is analyzed and calculated by the processor. Processing, so as to make decisions and control the execution system to realize vehicle acceleration, deceleration and steering.

The perception system is the eyes of unmanned vehicles, and the safety of autonomous driving largely depends on how well the sensing systems are doing. Monocular cameras are commonly used, cost-disclosed sensing elements in vehicles. However, due to the limitation of the monocular imaging principle, in the prior art, the obstacle distance cannot be obtained by the monocular camera alone, and the size and direction of the obstacle must be known, or the distance of the obstacle can only be obtained by using the binocular camera.

Contents of the invention

To this end, an embodiment of the present invention proposes a method for measuring obstacle distance based on a high-precision map and a monocular camera, to solve the problem that the existing technology cannot obtain the obstacle distance alone through the monocular camera, and the size and size of the obstacle must be known. Orientation, or the technical problem of using a binocular camera to obtain the distance of obstacles.

A method for measuring obstacle distances based on a high-precision map and a monocular camera according to an embodiment of the present invention includes:

Identify elements on the HD map;

Obtain the image captured by the monocular camera on the vehicle, and perform matching analysis on the obstacles in the image captured by the monocular camera and the recognized elements on the high-precision map;

Determine whether the height information of the obstacle is known;

If the height information of the obstacle is known, the distance between the obstacle and the vehicle is obtained based on the principle of monocular ranging and the height information of the obstacle.

According to the method for measuring obstacle distance based on a high-precision map and a monocular camera according to an embodiment of the present invention, firstly identify elements on the high-precision map, then acquire images captured by the monocular camera on the vehicle, and convert the images captured by the monocular camera to The obstacles in the system are matched with the elements on the recognized high-precision map. If the height information of the obstacles is known, the distance between the obstacles and the vehicle can be directly obtained based on the principle of monocular distance measurement and the height information of the obstacles. distance, the present invention only needs to use a monocular camera, does not need to use a binocular camera, and does not need to know the size and direction of the obstacle, and can directly measure the distance of the obstacle, and the implementation cost is low.

In addition, the method for measuring obstacle distance based on a high-precision map and a monocular camera according to the above-mentioned embodiments of the present invention may also have the following additional technical features:

Further, after the step of confirming whether the height information of the obstacle is known, the method further includes:

If the height information of the obstacle is unknown, judge whether the obstacle in the image captured by the monocular camera matches the elements on the recognized high-precision map;

If the obstacle in the image captured by the monocular camera matches the recognized elements on the high-precision map, the distance between the obstacle and the vehicle is obtained based on the principle of monocular ranging and the information of the high-precision map.

Further, after the step of judging whether the obstacle in the image captured by the monocular camera matches the recognized element on the high-precision map, the method further includes:

If the obstacle in the image taken by the monocular camera does not match all the elements on the recognized high-precision map, it is further judged whether there is a lane line;

If there is a lane line, the distance between the obstacle and the vehicle is calculated through the lane line.

Further, the step of calculating the distance between the obstacle and the vehicle through the lane line specifically includes:

Obtain the width of the lane from the high-precision map, and then calculate the distance of the obstacle and lane line calculation through the pixel points of the lane in the image, the width of the lane, the camera parameters of the monocular camera, and the height pixels of the obstacle in the image The distance between the obstacle and the vehicle.

Further, after the step of judging whether there is a lane line, the method further includes:

If there is no lane line, it is judged whether there is a preset road element, and the preset road element is any one of road markings, pedestrian crossings, and road poles;

If there are preset road elements, the perspective view is converted into a top view through inverse perspective transformation, and then interpolated to determine the lane line and calculate the distance between the obstacle and the vehicle.

Another embodiment of the present invention proposes a system for measuring obstacle distances based on a high-precision map and a monocular camera to solve the problem that the existing technology cannot obtain the obstacle distance alone through the monocular camera, and the size and direction of the obstacle must be known. Or the technical problem of using binocular cameras to obtain the distance of obstacles.

According to an embodiment of the present invention, a system for measuring obstacle distance based on a high-precision map and a monocular camera, the system includes:

The recognition module is used to recognize elements on the high-precision map;

The acquisition module is used to acquire the image taken by the monocular camera on the vehicle, and perform matching analysis on the obstacles in the image taken by the monocular camera and the recognized elements on the high-precision map;

The first judging module is used to judge whether the height information of the obstacle is known;

The first calculation module is configured to obtain the distance between the obstacle and the vehicle based on the principle of monocular ranging and the height information of the obstacle if the height information of the obstacle is known.

According to the system for measuring the distance of obstacles based on a high-precision map and a monocular camera according to an embodiment of the present invention, the elements on the high-precision map are first identified, and then the image captured by the monocular camera on the vehicle is acquired, and the image captured by the monocular camera is The obstacles in the system are matched with the elements on the recognized high-precision map. If the height information of the obstacles is known, the distance between the obstacles and the vehicle can be directly obtained based on the principle of monocular distance measurement and the height information of the obstacles. distance, the present invention only needs to use a monocular camera, does not need to use a binocular camera, and does not need to know the size and direction of the obstacle, and can directly measure the distance of the obstacle, and the implementation cost is low.

In addition, the system for measuring obstacle distances based on high-precision maps and monocular cameras according to the above-mentioned embodiments of the present invention may also have the following additional technical features:

Further, the system also includes:

The second judging module is used to judge whether the obstacle in the image captured by the monocular camera matches the elements on the recognized high-precision map if the height information of the obstacle is unknown;

The second calculation module is used to obtain the distance between the obstacle and the vehicle based on the principle of monocular ranging and the information of the high-precision map if the obstacle in the image captured by the monocular camera matches the recognized element on the high-precision map. distance between.

Further, the system also includes:

The third judging module is used to further judge whether there is a lane line if the obstacle in the image captured by the monocular camera does not match all the elements on the recognized high-precision map;

The third calculation module is used to calculate the distance between the obstacle and the vehicle through the lane line if there is a lane line.

Further, the third calculation module is specifically used for:

Obtain the width of the lane from the high-precision map, and then calculate the distance of the obstacle and lane line calculation through the pixel points of the lane in the image, the width of the lane, the camera parameters of the monocular camera, and the height pixels of the obstacle in the image The distance between the obstacle and the vehicle.

Further, the system also includes:

The fourth judging module is used to judge whether there is a preset road element if there is no lane line, and the preset road element is any one of road markings, pedestrian crossings, and road poles;

The fourth calculation module is used to convert the perspective view into a top view through inverse perspective transformation if there is a preset road element, and then perform interpolation changes to determine the lane line and calculate the distance between the obstacle and the vehicle.

Description of drawings

The above and/or additional aspects and advantages of the embodiments of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, wherein:

Fig. 1 is a schematic flowchart of a method for measuring obstacle distance based on a high-precision map and a monocular camera according to an embodiment of the present invention;

Fig. 2 is a schematic diagram of obstacles on the lane;

Figure 3 is a schematic diagram of the principle of calculating the height of the obstacle and the distance from the obstacle to the camera through similar triangles;

Fig. 4 is a schematic structural diagram of a system for measuring obstacle distances based on a high-definition map and a monocular camera according to an embodiment of the present invention.

Detailed ways

In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments It is a part of embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

Please refer to Fig. 1, a method for measuring obstacle distance based on a high-precision map and a monocular camera proposed by an embodiment of the present invention, the method includes the following steps S1 to S10:

S1, identify elements on the HD map.

S2. Obtain the image captured by the monocular camera on the vehicle, and perform matching analysis on the obstacles in the image captured by the monocular camera and the recognized elements on the high-precision map.

S3, judging whether the height information of the obstacle is known.

S4. If the height information of the obstacle is known, the distance between the obstacle and the vehicle is obtained based on the monocular ranging principle and the height information of the obstacle.

The height of obstacles can be used to assist distance measurement. When there is no map element, it can be measured by height. If you know the real height of the object and the pixels on the image, you can directly calculate the distance between the obstacle and the vehicle.

In this embodiment, after the step of confirming whether the height information of the obstacle is known, the method further includes:

S5, if the height information of the obstacle is unknown, judge whether the obstacle in the image captured by the monocular camera matches the elements on the recognized high-precision map;

S6, if the obstacle in the image captured by the monocular camera matches the recognized elements on the high-precision map, then based on the principle of monocular ranging and the information of the high-precision map, the distance between the obstacle and the vehicle is obtained.

In this embodiment, after the step of judging whether the obstacle in the image captured by the monocular camera matches the recognized element on the high-precision map, the method further includes:

S7, if the obstacle in the image captured by the monocular camera does not match all the elements on the recognized high-precision map, further determine whether there is a lane line;

S8. If there is a lane line, calculate the distance between the obstacle and the vehicle through the lane line.

In this embodiment, the step of calculating the distance between the obstacle and the vehicle through the lane line specifically includes:

Obtain the width of the lane from the high-precision map, and then calculate the distance of the obstacle and lane line calculation through the pixel points of the lane in the image, the width of the lane, the camera parameters of the monocular camera, and the height pixels of the obstacle in the image The distance between the obstacle and the vehicle.

In this embodiment, after the step of judging whether there is a lane line, the method further includes:

S9, if there is no lane line, it is judged whether there is a preset road element, and the preset road element is any one of road markings, pedestrian crossings, and road poles;

S10, if there is a preset road element, convert the perspective view into a top view through inverse perspective transformation, and then perform interpolation changes to determine the lane line and calculate the distance between the obstacle and the vehicle.

Specifically, if the obstacle is on the lane, make a line perpendicular to the lane line through the intersection point of the obstacle and the ground, as shown in Figure 2, then the calculation of the distance from the obstacle to the camera can be converted into the distance from the point to the camera. Assuming that the lane width of the lane is the same, and the lane width is obtained through the high-precision map, the height of the obstacle and the distance from the obstacle to the camera can be calculated through similar triangles, as shown in Figure 3. Assuming that the height of the obstacle remains unchanged, the height of the obstacle can be continuously optimized later. When there is no high-precision map to match the landmarks, the height of the obstacle can be used to calculate the distance through similar triangles.

Assuming that the two lane lines are parallel and equal in width, the width of the current lane can be obtained from the map, and then the obstacle can be calculated by the pixel point of the lane in the image, the width of the lane, the camera internal reference, and the height pixel of the obstacle in the image object distance and height. Among them, the obstacle is projected to the lane line, through the small hole imaging model, according to the similar triangle, let B, P' be the projection point on the pixel, and X' can be known. X can be obtained through the map, the focal length is known, and the distance Z between the obstacle and the vehicle can be calculated.

If the obstacle is an element on the HD map, its original size can be obtained directly through the HD map and the distance to it can be obtained through the camera projection principle

If the obstacle is on the pedestrian crossing/lane marking line, the obstacle can be obtained more accurately through the high-precision map. The perspective view is transformed into a top view through inverse perspective transformation, and then interpolated to determine the lane line calculation Obstacles and vehicles the distance between.

Taking the road pole as an example, the actual distance between the road pole and the camera is known, and the distance between the road pole and the camera and the distance between the obstacle and the camera are known on the image after the inverse projection transformation, and the distance between the obstacle and the vehicle can be calculated .

To sum up, according to the system for measuring obstacle distance based on high-precision map and monocular camera provided in this embodiment, the elements on the high-precision map are firstly identified, and then the image captured by the monocular camera on the vehicle is obtained, and the monocular camera The obstacles in the captured image are matched with the elements on the recognized high-precision map. If the height information of the obstacle is known, the obstacle and the vehicle can be directly obtained based on the principle of monocular distance measurement and the height information of the obstacle. The distance between obstacles, the present invention only needs to use a monocular camera, does not need to use a binocular camera, and does not need to know the size and direction of the obstacle, it can directly measure the distance of the obstacle, and the implementation cost is low.

Please refer to FIG. 4 , another embodiment of the present invention proposes a system for measuring obstacle distances based on a high-precision map and a monocular camera. The system includes:

The recognition module is used to recognize elements on the high-precision map;

The acquisition module is used to acquire the image taken by the monocular camera on the vehicle, and perform matching analysis on the obstacles in the image taken by the monocular camera and the recognized elements on the high-precision map;

The first judging module is used to judge whether the height information of the obstacle is known;

The first calculation module is configured to obtain the distance between the obstacle and the vehicle based on the principle of monocular ranging and the height information of the obstacle if the height information of the obstacle is known.

In this embodiment, the system also includes:

The second judging module is used to judge whether the obstacle in the image captured by the monocular camera matches the elements on the recognized high-precision map if the height information of the obstacle is unknown;

The second calculation module is used to obtain the distance between the obstacle and the vehicle based on the principle of monocular ranging and the information of the high-precision map if the obstacle in the image captured by the monocular camera matches the recognized element on the high-precision map. distance between.

In this embodiment, the system also includes:

The third judging module is used to further judge whether there is a lane line if the obstacle in the image captured by the monocular camera does not match all the elements on the recognized high-precision map;

The third calculation module is used to calculate the distance between the obstacle and the vehicle through the lane line if there is a lane line.

In this embodiment, the third calculation module is specifically used for:

Obtain the width of the lane from the high-precision map, and then calculate the distance of the obstacle and lane line calculation through the pixel points of the lane in the image, the width of the lane, the camera parameters of the monocular camera, and the height pixels of the obstacle in the image The distance between the obstacle and the vehicle.

In this embodiment, the system also includes:

The fourth judging module is used to judge whether there is a preset road element if there is no lane line, and the preset road element is any one of road markings, pedestrian crossings, and road poles;

The fourth calculation module is used to convert the perspective view into a top view through inverse perspective transformation if there is a preset road element, and then perform interpolation changes to determine the lane line and calculate the distance between the obstacle and the vehicle.

According to the system for measuring the obstacle distance based on the high-precision map and the monocular camera provided in this embodiment, the elements on the high-precision map are first identified, and then the image captured by the monocular camera on the vehicle is acquired, and the image captured by the monocular camera is The obstacles in the system are matched with the elements on the recognized high-precision map. If the height information of the obstacles is known, the distance between the obstacles and the vehicle can be directly obtained based on the principle of monocular distance measurement and the height information of the obstacles. distance, the present invention only needs to use a monocular camera, does not need to use a binocular camera, and does not need to know the size and direction of the obstacle, and can directly measure the distance of the obstacle, and the implementation cost is low.

The logic and/or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced listing of executable instructions for implementing logical functions, can be embodied in any computer-readable medium, For use with instruction execution systems, devices, or devices (such as computer-based systems, systems including processors, or other systems that can fetch instructions from instruction execution systems, devices, or devices and execute instructions), or in conjunction with these instruction execution systems, devices or equipment used. For the purposes of this specification, a "computer-readable medium" may be any device that can contain, store, communicate, propagate or transmit a program for use in or in conjunction with an instruction execution system, device or device.

More specific examples (non-exhaustive list) of computer-readable media include the following: electrical connection with one or more wires (electronic device), portable computer disk case (magnetic device), random access memory (RAM), Read Only Memory (ROM), Erasable and Editable Read Only Memory (EPROM or Flash Memory), Fiber Optic Devices, and Portable Compact Disc Read Only Memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program can be printed, as it may be possible, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or other suitable processing if necessary. The program is processed electronically and stored in computer memory.

It should be understood that various parts of the present invention can be realized by hardware, software, firmware or their combination. In the embodiments described above, various steps or methods may be implemented by software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented by any one or combination of the following techniques known in the art: Discrete logic circuits, ASICs with suitable combinational logic gates, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

In the description of this specification, descriptions referring to the terms "one embodiment", "some embodiments", "example", "specific examples", or "some examples" mean that specific features described in connection with the embodiment or example , structure, material or characteristic is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

Although the embodiments of the present invention have been shown and described, those skilled in the art can understand that various changes, modifications, substitutions and modifications can be made to these embodiments without departing from the principle and spirit of the present invention. The scope of the invention is defined by the claims and their equivalents.

Claims (10)

1. A method for measuring obstacle distance based on a high-precision map and a monocular camera, characterized in that, comprising: Identify elements on the HD map; Obtain the image captured by the monocular camera on the vehicle, and perform matching analysis on the obstacles in the image captured by the monocular camera and the recognized elements on the high-precision map; Determine whether the height information of the obstacle is known; If the height information of the obstacle is known, the distance between the obstacle and the vehicle is obtained based on the principle of monocular ranging and the height information of the obstacle. 2. The method for measuring obstacle distance based on high-precision map and monocular camera according to claim 1, characterized in that, after confirming whether the height information of the obstacle is known, the method also includes: If the height information of the obstacle is unknown, judge whether the obstacle in the image captured by the monocular camera matches the elements on the recognized high-precision map; If the obstacle in the image captured by the monocular camera matches the recognized elements on the high-precision map, the distance between the obstacle and the vehicle is obtained based on the principle of monocular ranging and the information of the high-precision map. 3. the method for measuring obstacle distance based on high-precision map and monocular camera according to claim 2, is characterized in that, judges whether the obstacle in the image that monocular camera takes and the element on the identified high-precision map After the step of matching, the method also includes: If the obstacle in the image taken by the monocular camera does not match all the elements on the recognized high-precision map, it is further judged whether there is a lane line; If there is a lane line, the distance between the obstacle and the vehicle is calculated through the lane line. 4. The method for measuring obstacle distance based on a high-precision map and a monocular camera according to claim 3, wherein the step of calculating the distance between the obstacle and the vehicle through the lane line specifically includes: Obtain the width of the lane from the high-precision map, and then calculate the distance of the obstacle and lane line calculation through the pixel points of the lane in the image, the width of the lane, the camera parameters of the monocular camera, and the height pixels of the obstacle in the image The distance between the obstacle and the vehicle. 5. the method for measuring obstacle distance based on high-precision map and monocular camera according to claim 3, is characterized in that, after the step of judging whether there is lane line, described method also comprises: If there is no lane line, it is judged whether there is a preset road element, and the preset road element is any one of road markings, pedestrian crossings, and road poles; If there are preset road elements, the perspective view is converted into a top view through inverse perspective transformation, and then interpolated to determine the lane line and calculate the distance between the obstacle and the vehicle. 6. A system for measuring obstacle distance based on a high-precision map and a monocular camera, characterized in that it includes: The recognition module is used to recognize elements on the high-precision map; The acquisition module is used to acquire the image taken by the monocular camera on the vehicle, and perform matching analysis on the obstacles in the image taken by the monocular camera and the recognized elements on the high-precision map; The first judging module is used to judge whether the height information of the obstacle is known; The first calculation module is configured to obtain the distance between the obstacle and the vehicle based on the principle of monocular ranging and the height information of the obstacle if the height information of the obstacle is known. 7. The system for measuring obstacle distance based on high-precision map and monocular camera according to claim 6, characterized in that, the system also includes: The second judging module is used to judge whether the obstacle in the image captured by the monocular camera matches the elements on the recognized high-precision map if the height information of the obstacle is unknown; The second calculation module is used to obtain the distance between the obstacle and the vehicle based on the principle of monocular ranging and the information of the high-precision map if the obstacle in the image captured by the monocular camera matches the recognized element on the high-precision map. distance between. 8. The system for measuring obstacle distance based on high-precision map and monocular camera according to claim 7, characterized in that, the system also includes: The third judging module is used to further judge whether there is a lane line if the obstacle in the image captured by the monocular camera does not match all the elements on the recognized high-precision map; The third calculation module is used to calculate the distance between the obstacle and the vehicle through the lane line if there is a lane line. 9. The method for measuring obstacle distance based on a high-precision map and a monocular camera according to claim 8, wherein the third calculation module is specifically used for: Obtain the width of the lane from the high-precision map, and then calculate the distance of the obstacle and lane line calculation through the pixel points of the lane in the image, the width of the lane, the camera parameters of the monocular camera, and the height pixels of the obstacle in the image The distance between the obstacle and the vehicle. 10. the system based on high precision map and monocular camera measuring obstacle distance according to claim 8, is characterized in that, described system also comprises: The fourth judging module is used to judge whether there is a preset road element if there is no lane line, and the preset road element is any one of road markings, pedestrian crossings, and road poles; The fourth calculation module is used to convert the perspective view into a top view through inverse perspective transformation if there is a preset road element, and then perform interpolation changes to determine the lane line and calculate the distance between the obstacle and the vehicle.

Images