Vehicle reverse driving detection method and device and computer readable storage medium

Abstract

The present invention discloses a vehicle reverse driving detection method and device and a computer-readable storage medium. The detection method comprises: obtaining a road image in a detection area, determining the probability that a lane in the detection area is a forward lane based on a neural network model; obtaining images of several moving vehicles in the detection area, determining the reverse driving probability of several vehicles based on a neural network model; and determining whether several vehicles have reverse driving behavior according to the probability that the lane is a forward lane and the reverse driving probability of several vehicles. When determining whether a vehicle has reverse driving behavior, the present invention determines the confidence of this result after obtaining the result of the lane line, and accordingly, confirms the confidence of this result after obtaining the result that the vehicle is in the reverse direction. Based on the confidence of the above two results, it is determined that the vehicle has reverse driving behavior, so that the detection method of the present invention can refer to more basic information to help judge, so that the detection result has higher robustness.

Description

Technical Field

[0001] The present invention belongs to the technical field of wrong-way detection, and in particular, relates to a vehicle wrong-way detection method and device and a computer-readable storage medium. Background Art

[0002] With the rapid development of my country's social economy, the continuous expansion of population and the acceleration of urbanization, the number of motor vehicles in my country is also increasing rapidly, which has caused more problems of motor vehicle violations.

[0003] Most existing vehicle wrong-way detection methods are based on fixed scenes, that is, the camera is set up at the checkpoint or above the road, the shooting area is relatively fixed, and the scope of action is limited. At the same time, existing methods can often only identify wrong-way behavior in the current lane, but cannot identify wrong-way behavior on surrounding roads.

[0004] The patent application with application number CN201811654238.0 provides a method, device and system for monitoring vehicle violations. The method includes: obtaining a current monitoring image; performing traffic sign recognition on the current monitoring image to obtain the category of the traffic sign contained in the current monitoring image; performing vehicle recognition on the current monitoring image to obtain the information of the vehicle contained in the current monitoring image; and judging whether the vehicle is in violation of the traffic rules based on the category of the traffic sign and the information of the vehicle.

[0005] The above scheme adopts a judgment method of either "no" or "yes", and the accuracy and credibility of the judgment results cannot meet the requirements of the existing market.

[0006] In view of this, the present invention is proposed. Summary of the invention

[0007] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a vehicle reverse driving detection method. By obtaining various basic algorithm results related to reverse driving, and then establishing reverse driving judgment logic, it is determined whether the vehicle is reverse driving. In this way, the detection method of the present invention can refer to more basic information to help judge, so that the detection result has higher robustness.

[0008] Another object of the present invention is to provide a vehicle reverse driving detection device.

[0009] Another object of the present invention is to provide a computer-readable medium for implementing the above-mentioned vehicle wrong-way detection method.

[0010] In order to solve the above technical problems, the basic concept of the technical solution adopted by the present invention is:

[0011] A vehicle reverse driving detection method comprises the following steps:

[0012] Obtain a road image within the detection area, and determine the probability that the lane within the detection area is a forward lane based on a neural network model;

[0013] Acquire images of several moving vehicles in the detection area, and determine the reverse driving probabilities of the several vehicles based on the neural network model;

[0014] According to the probability that the lane is a forward lane and the reverse driving probabilities of the plurality of vehicles, it is determined whether the plurality of vehicles have reverse driving behaviors.

[0015] Further, the method of acquiring the road image in the detection area and acquiring the images of several moving vehicles in the detection area is:

[0016] The road information in the detection area is collected in a mobile manner and images are generated while moving.

[0017] Further, the method for determining that the lane in the detection area is a forward lane includes the following steps:

[0018] Determine that the lane where the mobile data collection equipment is located is the forward lane;

[0019] It is determined whether the lane line in the road image is a dotted line. If so, the lanes on both sides of the lane line are determined to be forward lanes.

[0020] Furthermore, the method for determining the reverse driving probability of a plurality of vehicles includes: obtaining a moving target map by a moving target detection algorithm, denoted as M, and denoting the bounding boxes of all target vehicles detected in the current frame as Cn, where n is 1, 2, 3, ...;

[0021] From this we can get the moving vehicle On, On=M☉Cn,

[0022] Where ☉ is the cutout operation of the bounding box on the image;

[0023] The obtained moving vehicle image is input into the trained reverse driving detection model to obtain the reverse driving probability of the vehicle.

[0024] Furthermore, determining whether a number of vehicles have reverse driving behavior includes determining the probability that a target vehicle among the number of vehicles has reverse driving behavior, and determining whether the target vehicle has reverse driving behavior based on the probability that the target vehicle has reverse driving behavior.

[0025] Furthermore, the probability of determining that the target vehicle is driving in the opposite direction is obtained by formula (1):

[0026] Formula (1):

[0027] Among them, P(R) is the probability that the lane line is a dotted line, P(D) is the probability that the target vehicle is facing the opposite direction, and P(Dn ') is the probability that the nth vehicle in the lane where the target vehicle is located is facing the opposite direction except the target vehicle; a 1 is the weighted coefficient of the probability that the lane line is a dotted line, a 2 is the weighting coefficient of the probability that the target vehicle is facing the opposite direction.

[0028] Furthermore, after the mobile collection of road information in the detection area and the generation of images while moving, the following steps are also included: scene semantic segmentation of the road image in the detection area.

[0029] A vehicle reverse running detection device, comprising:

[0030] A data collection module, used for mobile collection of road information and generating data while moving;

[0031] A first determination module, for determining a probability that the lane is a forward lane based on the data of the road information;

[0032] A second determination module is used to determine the probability that the vehicle is traveling in the wrong direction based on the road information data;

[0033] The third determination module is used to determine whether the vehicle has reverse driving behavior based on the probability that the lane is a forward lane and the probability that the vehicle is driving in the reverse direction.

[0034] Furthermore, the data acquisition module includes:

[0035] A mobile unit for moving on roads;

[0036] The camera unit is connected to the mobile unit and is used to collect images of vehicles and lane lines within the field of view.

[0037] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the vehicle wrong-way detection method as described above are executed.

[0038] After adopting the above technical scheme, the present invention has the following beneficial effects compared with the prior art.

[0039] 1. The present invention is different from the traditional method of capturing illegal behaviors in fixed scenes. It is more difficult to judge whether a vehicle is driving in the wrong direction in a mobile camera scene. The solution proposed in the present invention solves this problem to a certain extent.

[0040] 2. The present invention proposes to first identify various basic information in the road scene, and then solve the problem of detecting vehicles driving against traffic regulations by establishing a scene-inspired model.

[0041] 3. The present invention constructs a scene model based on the results of various basic algorithms to make a reverse driving judgment on the vehicle in the lane. The present invention not only utilizes the object vehicle's own information, but also utilizes the scene information around the vehicle.

[0042] The specific implementation modes of the present invention are further described in detail below in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The accompanying drawings are part of the present invention and are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but do not constitute an improper limitation of the present invention. Obviously, the drawings described below are only some embodiments. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the accompanying drawings:

[0044] Figure 1 It is a flow chart of a vehicle reverse driving detection method of the present invention;

[0045] Figure 2 It is a schematic diagram of a vehicle reverse driving detection device of the present invention.

[0046] It should be noted that these drawings and textual descriptions are not intended to limit the conceptual scope of the present invention in any way, but are intended to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The following embodiments are used to illustrate the present invention but are not used to limit the scope of the present invention.

[0048] In the description of the present invention, it should be noted that the directions or positional relationships indicated by terms such as “upper”, “lower”, “front”, “back”, “left”, “right”, “vertical”, “inside” and “outside” are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as a limitation on the present invention.

[0049] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0050] like Figure 1 and Figure 2 As shown, the present invention provides a vehicle wrong-way detection method, device and computer-readable medium.

[0051] Among them, Figure 1 As shown, the vehicle reverse driving detection method includes the following steps: acquiring a road image within a detection area, and determining the probability that a lane in the detection area is a forward lane based on a neural network model; acquiring images of several moving vehicles in the detection area, and determining the reverse driving probability of several vehicles based on a neural network model; and determining whether several vehicles have reverse driving behavior based on the probability that the lane is a forward lane and the reverse driving probability of several vehicles.

[0052] In detail, due to the complex and changeable real environment, the lane line detection algorithm, vehicle position detection algorithm, and vehicle direction detection algorithm are easily affected by external factors such as weather, lighting, and occlusion. Therefore, the algorithm detection results are often different from the actual situation. The judgment of wrong traffic violations requires a comprehensive judgment using the results of algorithms such as lane lines and vehicle directions. If one of the algorithms has an erroneous result, it will have a direct impact on the final judgment. Therefore, the above-mentioned wrong detection method provided by the present invention adds artificial confidence on the basis of obtaining the results of various elements for judging the vehicle's wrong traffic to alleviate this phenomenon.

[0053] In other words, the present invention is different from the prior art in that it directly uses the algorithm to determine whether the vehicle is driving in the wrong direction. Instead, the present invention determines the confidence of the result after obtaining the lane line result. Correspondingly, after obtaining the result that the vehicle is heading in the wrong direction, the confidence of this result is also confirmed. Based on the confidence of the above two results, it is determined that the vehicle is driving in the wrong direction, so that the detection method of the present invention can refer to more basic information to help judge, making the detection result more robust.

[0054] In an optional embodiment of the present invention, a method for acquiring a road image within a detection area and an image of several moving vehicles within the detection area is: mobile acquisition of road information within the detection area and generation of images while moving.

[0055] In detail, a mobile camera is used to obtain video images of the road. The camera can be set up in front of or behind a moving vehicle or robot to provide a video stream for the algorithm.

[0056] Based on this, the method of the present invention for determining that a lane in a detection area is a forward lane includes the following steps: determining that the lane where the mobile data collection device is located is the forward lane; judging whether the lane line in the road image is a dotted line, and if so, determining that the lanes on both sides of the lane line are the forward lanes.

[0057] In detail, based on the video image of the road obtained by the mobile camera, the lane line detection algorithm SCNN is used to obtain the lane line result, and the lane line detection algorithm result is recorded as L, where L represents the curve equation of the lane line in the image coordinate system. Left is the type of the left lane line, and Right is the type of the right lane line. Assuming that n lane lines can be detected in one frame of image, the method for determining the lane driving direction according to the output result of the lane line is as follows:

[0058] 1. By default, the current lane of the vehicle where the mobile camera is located is the forward lane;

[0059] 2. If the lane line on the left or right side of the current lane is a dotted line, the lanes on the left and right sides of the lane are also considered to be forward lanes;

[0060] 3. If the left or right lane line of the forward lane determined in step 2 is a dotted line, then the left lane or right lane of the lane is also determined to be the forward lane.

[0061] Then we can define the set of forward lanes R = {R1, R2, R3, ...}. R1 is the lane, and Rn (n>=2) can be expressed as:

[0062] Rn = {Left(R*) is a dotted line, R* is the left lane or Right(R*) is a dotted line, R* is the right lane};

[0063] Where R* belongs to the set R.

[0064] Furthermore, in an embodiment of the present invention, a method for determining the probability of a number of vehicles traveling in the opposite direction includes: obtaining a moving target map through a moving target detection algorithm YoloV3, denoted as M, and denoting the bounding boxes of all target vehicles detected in the current frame as Cn, where n is 1, 2, 3...; thereby obtaining a moving vehicle On, On = M☉Cn, where ☉ is a cutout operation of the bounding box on the image; and inputting the obtained moving vehicle image into a trained reverse detection model AlexNet to obtain the reverse driving probability of the vehicle.

[0065] The present invention determines whether a plurality of vehicles have reverse driving behavior, including determining the probability that a certain target vehicle among the plurality of vehicles has reverse driving behavior, and determining whether the target vehicle has reverse driving behavior according to the probability that the target vehicle has reverse driving behavior.

[0066] In detail, the present invention does not directly obtain a "yes" or "no" conclusion when detecting whether a vehicle is driving in the wrong direction, but determines whether the vehicle is a wrong-way illegal vehicle by obtaining the probability that a certain moving vehicle is driving in the wrong direction.

[0067] Specifically, after obtaining the moving vehicle set On and the front lane set Rn, we start to determine the direction of the moving vehicles in the front lane and look for the vehicle whose moving vehicle direction in the front lane is determined to be the front of the vehicle, that is, the vehicle facing the front of the vehicle. Assume that the reverse vehicles are represented by the set V.

[0068] It can be seen that the present invention also uses the driving information of other vehicles in the scene as a basis for judgment, not only understanding the object itself, but also understanding the surrounding environment of the object and making a comprehensive judgment, which further improves the accuracy of judgment.

[0069] The probability of determining that the target vehicle is driving in the opposite direction is obtained by formula (1):

[0070] Formula (1):

[0071] Among them, P(R) is the probability that the lane line is a dotted line, P(D) is the probability that the target vehicle is facing the opposite direction, and P(D n ') is the probability that the nth vehicle in the lane where the target vehicle is located is facing the opposite direction except the target vehicle; a 1 is the weighted coefficient of the probability that the lane line is a dotted line, a 2 is the weighting coefficient of the probability that the target vehicle is facing the opposite direction.

[0072] In the above scheme, P(R) represents the confidence that the lane line is a dotted line, and P(D) represents the confidence that the moving vehicle is in the opposite direction. P(R) and P(D) are obtained by the SCNN algorithm and the Alexnet algorithm respectively. 1 Represents the weighting coefficient of the dashed line confidence, a 2 is the weighted coefficient of the confidence level of the illegal vehicle driving in the wrong direction. The weighted coefficient can be a value verified by experiments. The weighted coefficient is set in advance according to the actual situation. Here a 1 Take 0.5, a 2 Take 0.2.

[0073] P(Dn') represents the probability of other vehicles in the lane where the target vehicle is detected to be driving in the opposite direction. Therefore, the last term in formula (1) represents the average probability of all other vehicles in the lane where the target vehicle is driving in the opposite direction. If the probability is large, it means that the current target vehicle is likely to be misjudged. This is in line with the concept of scene inspiration.

[0074] In some embodiments of the present invention, after the mobile collection of road information in the detection area and the generation of images while moving, the following steps are further included: scene semantic segmentation of the road image in the detection area. For example, moving vehicles, vehicle orientations, lane line types, etc. related to the elements for judging the reverse traffic are segmented. The above scheme can be performed using the scene semantic segmentation algorithm UNet.

[0075] The vehicle reverse detection device provided by the present invention is as follows: Figure 2 As shown, including:

[0076] A data collection module, used for mobile collection of road information and generating data while moving;

[0077] A first determination module, for determining a probability that the lane is a forward lane based on the data of the road information;

[0078] A second determination module is used to determine the probability that the vehicle is traveling in the wrong direction based on the road information data;

[0079] The third determination module is used to determine whether the vehicle has reverse driving behavior based on the probability that the lane is a forward lane and the probability that the vehicle is driving in the reverse direction.

[0080] Furthermore, the data acquisition module includes:

[0081] A mobile unit for moving on roads;

[0082] The camera unit is connected to the mobile unit and is used to collect images of vehicles and lane lines within the field of view.

[0083] In detail, the present invention is different from the traditional method of capturing illegal behaviors in fixed scenes. It is more difficult to judge whether a vehicle is driving in the wrong direction in a mobile camera scene. The solution proposed in the present invention solves this problem to a certain extent.

[0084] In addition, it is currently difficult to directly solve the problem of vehicle reversing behavior using methods based solely on traditional algorithms or deep learning. The present invention proposes to first identify various basic information in the road scene, and then solve the problem of reversing illegal driving by establishing a scene-inspired model.

[0085] Moreover, the present invention constructs a scene model based on the results of various basic algorithms to make a reverse driving judgment on the vehicle on the lane. The present invention not only utilizes the object vehicle's own information, but also utilizes the scene information around the vehicle.

[0086] Corresponding to the above-mentioned vehicle wrong-way detection method, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned vehicle wrong-way detection method are executed.

[0087] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the vehicle reverse detection device described above can refer to the corresponding process in the method embodiment, and will not be repeated in the present invention. In the several embodiments provided by the present invention, it should be understood that the disclosed system, device and method can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0088] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0089] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0090] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, ROM, RAM, magnetic disks, or optical disks.

[0091] The above is only a preferred embodiment of the present invention, and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment, it is not used to limit the present invention. Any technician familiar with this patent can make some changes or modify the technical contents suggested above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of the solution of the present invention.

Claims

1. A vehicle reverse driving detection method, characterized in that: The following steps are involved: Obtain a road image within the detection area, and determine the probability that the lane within the detection area is a forward lane based on a neural network model; Acquire images of several moving vehicles in the detection area, and determine the reverse driving probabilities of the several vehicles based on the neural network model; According to the probability that the lane is a forward lane and the probability that the vehicles are traveling in the opposite direction, determining whether the vehicles are traveling in the opposite direction; Determining whether a number of vehicles are traveling in the wrong direction, including determining the probability that a target vehicle among the number of vehicles is traveling in the wrong direction, and determining whether the target vehicle is traveling in the wrong direction based on the probability that the target vehicle is traveling in the wrong direction; The probability of determining that the target vehicle is driving in the opposite direction is obtained by formula (1): Formula (1): Among them, P(R) is the probability that the lane line is a dotted line, P(D) is the probability that the target vehicle is facing the opposite direction, and P(D n ') is the probability that the nth vehicle on the lane where the target vehicle is located is facing the opposite direction except the target vehicle; a1 is the weighted coefficient of the probability that the lane line is a dotted line, and a2 is the weighted coefficient of the probability that the target vehicle is facing the opposite direction.

2. A vehicle reverse driving detection method according to claim 1, characterized in that: The method of acquiring the road image in the detection area and acquiring the images of several moving vehicles in the detection area is as follows: The road information in the detection area is collected in a mobile manner and images are generated while moving.

3. A vehicle reverse driving detection method according to claim 2, characterized in that: The method for determining that a lane in a detection area is a forward lane comprises the following steps: Determine that the lane where the mobile data collection equipment is located is the forward lane; It is determined whether the lane line in the road image is a dotted line. If so, the lanes on both sides of the lane line are determined to be forward lanes.

4. A vehicle reverse driving detection method according to claim 2, characterized in that: The method for determining the reverse driving probability of several vehicles includes: obtaining a moving target map by a moving target detection algorithm, denoted as M, and denoting the bounding boxes of all target vehicles detected in the current frame as Cn, where n is 1, 2, 3, ...; From this we can get the moving vehicle On, On=M☉Cn, Where ☉ is the cutout operation of the bounding box on the image; The obtained moving vehicle image is input into the trained wrong-way detection model to obtain the wrong-way driving probability of the vehicle.

5. A vehicle reverse driving detection method according to claim 2, characterized in that: After the road information in the detection area is collected in a mobile manner and an image is generated while moving, the following steps are also included: scene semantic segmentation is performed on the road image in the detection area.

6. A vehicle reverse detection device, characterized in that: For use in the vehicle reverse driving detection method according to any one of claims 1 to 5, comprising: A data collection module, used for mobile collection of road information and generating data while moving; A first determination module, for determining a probability that the lane is a forward lane based on the data of the road information; A second determination module is used to determine the probability that the vehicle is traveling in the wrong direction based on the road information data; The third determination module is used to determine whether the vehicle has reverse driving behavior based on the probability that the lane is a forward lane and the probability that the vehicle is driving in the reverse direction.

7. A vehicle reverse running detection device according to claim 6, characterized in that: The data acquisition module comprises: A mobile unit for moving on roads; The camera unit is connected to the mobile unit and is used to collect images of vehicles and lane lines within the field of view.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the vehicle wrong-way detection method according to any one of claims 1 to 5 are executed.

Images