CN116189430A

CN116189430A - Vehicle management method, device, equipment and storage medium for expressway entrance

Info

Publication number: CN116189430A
Application number: CN202310113667.1A
Authority: CN
Inventors: 应昶; 钟福祥; 杨云; 金泽民; 斯倩; 虞骏刚; 温称福; 吴涛涛
Original assignee: Zhejiang Jiaotou Expressway Operation Management Co ltd
Current assignee: Zhejiang Jiaotou Expressway Operation Management Co ltd
Priority date: 2023-02-07
Filing date: 2023-02-07
Publication date: 2023-05-30

Abstract

The embodiment of the invention discloses a vehicle management method, a device, equipment and a storage medium for a highway entrance, which relate to the technical field of highway cooperation, wherein a truck weighing area is arranged at a preset distance from a highway toll gate entrance, whether a vehicle to be identified has a casting event or not is determined by acquiring data to be identified of the truck weighing area, wherein the vehicle to be identified stays in the truck weighing area for a preset time period, if the vehicle to be identified has the casting event, a first prompt message is output to prompt the vehicle to be identified to drive into a preset safety area for rectifying goods loaded by the vehicle to be identified, and if the vehicle to be identified does not have the casting event, a second prompt message is output to prompt the vehicle to be identified to drive into the highway toll gate, so that the vehicle which is easy to induce the casting event can be found and excluded in time to enter the highway, and casting objects on the driving road surface of the highway are reduced, and the driving safety of the highway is ensured.

Description

Vehicle management method, device, equipment and storage medium for expressway entrance

Technical Field

The invention relates to the technical field of vehicle-road coordination, in particular to a vehicle management method, device, equipment and storage medium for an expressway entrance.

Background

Road sprinklers are one of the important causes of highway traffic accidents. For example, road throws tend to slow down the speed of vehicle traffic, thereby causing traffic accidents. In view of this, road throwing behavior is required to be recognized in time as a traffic event affecting running safety.

At present, the method mainly comprises the steps of detecting the casting objects on the driving pavement of the expressway in a manual inspection mode or a detection mode based on a video detection algorithm, and cleaning the casting objects on the pavement and giving a warning to a subsequent vehicle driving to the casting objects when the casting objects are detected. There are two periods from the discovery of a road throwing an obstacle to informing each driver that the obstacle will be passed, immediately from informing a professional road maintainer to clearing the throwing an obstacle. The existence of the two time periods can possibly cause traffic accidents and endanger the life and property safety of people.

Therefore, the method adopts a manual inspection mode or a detection mode based on a video detection algorithm to detect the casting objects on the driving road surface of the expressway so as to solve the safety risk of casting the obstacles on the expressway, and further prevent traffic accidents, which obviously cannot be achieved.

Disclosure of Invention

The embodiment of the invention provides a vehicle management method, device, equipment and storage medium for an expressway entrance, which are used for reducing casting matters on a driving road surface of an expressway and guaranteeing driving safety of the expressway.

In one aspect, an embodiment of the present invention provides a vehicle management method for an expressway entrance, the method including:

acquiring data to be identified of a truck weighing area after a vehicle to be identified stays in the truck weighing area for a preset time period; the truck weighing area is spaced from the entrance of the expressway toll station by a preset distance;

carrying out sprinkling event identification according to the data to be identified, and determining whether the sprinkling event occurs to the vehicle to be identified;

if the vehicle to be identified has a throwing event, outputting first prompt information to prompt the vehicle to be identified to drive into a preset safety area so as to rectify and change cargoes loaded by the vehicle to be identified;

and if the vehicle to be identified does not have a throwing event, outputting second prompting information to prompt the vehicle to be identified to drive into the entrance of the expressway toll station.

In another aspect, an embodiment of the present invention provides a vehicle management apparatus for an expressway entrance, the apparatus including:

The acquisition module is used for acquiring the data to be identified of the truck weighing area after the vehicle to be identified stays in the truck weighing area for a preset time; the truck weighing area is spaced from the entrance of the expressway toll station by a preset distance;

the identification module is used for carrying out sprinkling event identification according to the data to be identified and determining whether the sprinkling event occurs to the vehicle to be identified;

the first processing module is used for outputting first prompt information to prompt the vehicle to be identified to drive into a preset safety area to rectify cargoes loaded by the vehicle to be identified if the vehicle to be identified has a throwing event;

and the second processing module is used for outputting second prompt information to prompt the vehicle to be identified to drive into the entrance of the expressway toll station if the vehicle to be identified does not have a throwing event.

In another aspect, an embodiment of the present invention provides a vehicle management apparatus for an expressway portal, including a memory and a processor; the memory stores an application program, and the processor is configured to run the application program in the memory, so as to execute the steps in the vehicle management method for the highway entrance.

In another aspect, an embodiment of the present invention provides a storage medium storing a plurality of instructions adapted to be loaded by a processor to perform the steps in the method for managing vehicles at highway entrance described above.

The embodiment of the invention provides a vehicle management method, device, equipment and storage medium for a highway entrance, which relate to the technical field of highway cooperation, wherein a truck weighing area is arranged at a preset distance from a highway toll station entrance, and whether a vehicle to be identified has a casting event or not is determined by acquiring data to be identified of the truck weighing area, wherein the vehicle to be identified stays in the truck weighing area for a preset time period, according to the casting event identification of the data to be identified, if the vehicle to be identified has the casting event, a first prompt message is output to prompt the vehicle to be identified to drive into a preset safety area for rectifying goods loaded by the vehicle to be identified, and if the vehicle to be identified does not have the casting event, a second prompt message is output to prompt the vehicle to be identified to drive into the highway toll station entrance, so that the vehicle which is easy to induce the casting event can be discovered and excluded in time to enter the highway, and casting objects on the driving road surface of the highway are reduced, and the driving safety of the highway is ensured.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a schematic diagram of an application scenario of a vehicle management method for an expressway entrance according to an embodiment of the present invention;

FIG. 2 is a flow chart of a method for managing vehicles at an expressway entrance according to an embodiment of the invention;

FIG. 3 is a schematic view of a weight detection zone provided by an embodiment of the present invention;

FIG. 4 is a schematic diagram of a feature extraction network according to an embodiment of the present invention;

fig. 5 is a schematic structural view of a vehicle management device for highway entrance according to an embodiment of the present invention;

fig. 6 is a schematic structural diagram of a vehicle management apparatus for highway entrance according to an embodiment of the present invention.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to fall within the scope of the invention.

As described in the background art, the existing method mainly adopts a manual inspection mode or a detection mode based on a video detection algorithm to detect the casting objects on the driving road surface of the expressway, although the casting objects on the road surface can be identified, and the identified casting objects are clearly and warning to the following vehicles, so that the traffic safety accidents of the expressway are reduced. But there are two periods from the discovery of road throwing an obstacle to the notification of each driver who will pass the throwing an obstacle, and from the immediate notification of a professional road maintainer to the removal of the throwing an obstacle, the existence of both periods is also likely to cause a traffic accident. Although the prior art can shorten two time periods by improving the recognition efficiency so as to reduce the probability of possibly causing traffic accidents, hidden danger of causing traffic accidents still exists, and the vehicle on the expressway runs fast, so that the detection difficulty of detecting the casting objects on the running pavement of the expressway is increased to a certain extent, and the running safety of the expressway is difficult to be ensured.

Considering that vehicles need to pass through the entrance of the expressway toll gate before entering the expressway, the entrance of the expressway toll gate is an effective gate for controlling the loading problem of trucks, and vehicles which are easy to induce a casting event can be timely found and excluded from entering the expressway, but at present, the vehicles which are irregular in loading and easy to cast are difficult to monitor in place, the entrance of the expressway toll gate is narrow, and when the vehicles need to change the loading of cargoes, the vehicles are difficult to turn around, and traffic jam is possibly caused.

Based on the above, in order to reduce the casting of the driving road surface of the expressway and ensure the driving safety of the expressway, and at the same time, the embodiment of the invention provides a vehicle management method, device, equipment and storage medium for an expressway entrance, wherein a truck weighing area is arranged at a preset distance from the expressway toll gate entrance, the casting event identification is carried out according to the data to be identified by acquiring the data to be identified of the truck weighing area with the preset duration of time when the vehicle to be identified stays in the truck weighing area, if the vehicle to be identified has the casting event, a first prompt message is output to prompt the vehicle to be identified to drive into a preset safety area for rectifying the goods loaded by the vehicle to be identified, and if the vehicle to be identified does not have the casting event, a second prompt message is output to prompt the vehicle to be identified to drive into the expressway toll gate entrance, so that the vehicle which is easy to induce the casting event can be found and excluded from entering the expressway in time, and the driving safety of the expressway is reduced; and the throwing event identification is carried out on the data to be identified of the truck weighing area which is at a preset distance from the entrance of the expressway toll station, so that the probability of causing traffic jam can be reduced.

In order to facilitate understanding of the technical scheme of the present invention, the vehicle management method for the highway entrance provided by the present invention will be described below in connection with an actual application scenario.

As shown in fig. 1, fig. 1 is a schematic diagram of an application scenario of a vehicle management method for an expressway entrance provided by the embodiment of the invention, where the application scenario includes a truck weighing area, a vehicle to be identified, a vehicle management system for the expressway entrance, a preset first driving channel, a preset second driving channel, and a preset safety area, the vehicle management system for the expressway entrance is disposed in the truck weighing area, the preset first driving channel drives to the preset safety area, the preset second driving channel drives to an expressway toll gate entrance, the vehicle to be identified drives to the truck weighing area and stays for a preset time period, the vehicle management system for the expressway entrance performs a casting event identification according to the data to be identified of the truck weighing area where the vehicle to be identified stays for the preset time period in the truck weighing area, obtains a casting event identification result of the vehicle to be identified, and prompts the vehicle to be identified to drive into the preset safety area through the preset first driving channel according to the casting event identification result, or prompts the vehicle to be identified to drive into the expressway toll gate through the preset second driving channel, wherein whether the casting event identification result represents the vehicle to be identified.

Specifically, when the condition that the vehicle to be identified drives into the truck weighing area is monitored, the vehicle management system of the expressway entrance sends prompt information to prompt the vehicle to be identified to stay in the truck weighing area for a preset time period, the preset first driving channel and the preset second driving channel are closed, the vehicle management system of the expressway entrance obtains the data to be identified of the truck weighing area, the vehicle to be identified stays in the truck weighing area for the preset time period, the risk of throwing identification is carried out according to the data to be identified, whether the vehicle to be identified has a throwing event is determined, if the vehicle to be identified has the throwing event, the first prompt information is output, the preset first driving channel is opened, the cargo loaded by the vehicle to be identified is rectified and changed in the preset safety area through the preset first driving channel, if the vehicle to be identified does not have the throwing event, the second prompt information is output, the preset second driving channel is opened, and the vehicle to be identified drives into the expressway toll station entrance through the preset second driving channel. The first prompt information is used for prompting the vehicle to be identified to drive into a preset safety area and rectifying and modifying goods loaded on the vehicle to be identified, and the second prompt information is used for prompting the vehicle to be identified to drive into an entrance of a highway toll station.

As shown in fig. 1, the vehicle management system of the highway entrance includes a processing unit, a sensing unit, an output unit, a first valve, and a second valve. The sensing unit comprises a camera, an image sensor, a weight sensor, a pressure sensor, a gravity sensor, a radar and the like, wherein the first valve is arranged on a preset first running channel and used for controlling the opening and closing of the preset first running channel, and the second valve is arranged on a preset second running channel and used for controlling the opening and closing of the preset second running channel.

When a vehicle to be identified drives into a truck weighing area, a processing unit of a vehicle management system at an expressway entrance controls a first valve and a second valve to be closed, a sensing unit of the vehicle management system at the expressway entrance is started to acquire data to be identified of the truck weighing area, wherein the vehicle to be identified stays in the truck weighing area for a preset time period, and the data to be identified is transmitted to the processing unit; the processing unit carries out the casting risk identification according to the data to be identified, determines whether a casting event occurs on the vehicle to be identified, and when the casting event occurs on the vehicle to be identified, the processing unit controls the output unit to output first prompt information and controls the first valve to be opened so as to open a preset first running channel, and when the casting event does not occur on the vehicle to be identified, the processing unit controls the output unit to output second prompt information and controls the second valve to be opened so as to open a preset second running channel.

In some embodiments, the vehicle management system of the expressway entrance may further obtain a risk degree of throwing the vehicle to be identified according to the data to be identified of the truck weighing area, and when the risk degree of throwing the vehicle to be identified is greater than or equal to a preset risk degree threshold, the processing unit controls the output unit to output alarm information, and controls the first valve to open so as to open a preset first driving channel; when the vehicle to be identified is smaller than the preset risk degree threshold, the processing unit controls the second valve to open so as to open the preset second driving channel.

It should be noted that the above application scenario is only an exemplary illustration, and does not limit the vehicle management method for the highway entrance provided by the embodiment of the present invention.

As shown in fig. 2, fig. 2 is a flow chart of a vehicle management method for an expressway entrance according to an embodiment of the present invention, where the vehicle management method for an expressway entrance is applied to the vehicle management system for an expressway entrance shown in fig. 1, and the vehicle management system for an expressway entrance is deployed in a truck weighing area at a preset distance from an expressway toll gate entrance, and a vehicle to be identified stays for a preset period of time after being input into the truck weighing area. Specifically, the vehicle management method for the expressway entrance at least comprises the steps 210 to 240:

210, acquiring data to be identified of a truck weighing zone after the vehicle to be identified stays in the truck weighing zone for a preset time.

Wherein the data to be identified of the truck weighing zone includes, but is not limited to, the travel track of the vehicle to be identified, a sequence of vehicle images, a weighing zone image, and the like. The driving track can be the driving track of the vehicle to be identified which drives into the preset detection area, or the driving track of the vehicle to be identified in the detection range of the sensing unit in the vehicle management system of the expressway entrance, wherein the detection range of the sensing unit is larger than the preset detection area. The sequence of vehicle images may be successive frame images of the vehicle to be identified that is stationary in the truck weighing zone for a preset period of time. The weighing area image can be an image of a truck weighing area acquired after a truck to be identified exits the truck weighing area.

In an alternative embodiment, a sensing unit can be arranged in the truck weighing zone, the weight of the cargo of the vehicle to be identified is collected through a gravity sensor, a weight sensor or a pressure sensor in the sensing unit, and the driving track of the vehicle to be identified, which drives into the preset detection zone, is obtained through a camera, an image sensor or a radar in the sensing unit.

In an alternative embodiment, a target area is preset at a preset distance in front of a truck weighing area, a sensing unit is arranged in a road between the target area and the truck weighing area, when a vehicle is detected to drive into the target area in a target lane, the vehicle is set as a vehicle to be identified, prompt information is sent to prompt the vehicle to be identified to drive into the truck weighing area, and the sensing unit is started to acquire the weight of goods, carriage contour information, driving track and vehicle image sequence in the process that the vehicle to be identified drives into the truck weighing area from the target area, and when the vehicle to be identified is detected to drive out of the truck weighing area, the sensing unit is started to acquire images of the truck weighing area to acquire data to be identified of the truck weighing area.

Illustratively, taking the weight of the cargo as an example, when the weight of the cargo is the weight of the vehicle to be identified, as shown in fig. 3 (a), a gravity sensor, a weight sensor or a pressure sensor is disposed on the central line of the road between the target area and the truck weighing zone at intervals of a preset length to form a first weight detection zone, and when the vehicle to be identified moves from the target area to the truck weighing zone, the weight of the vehicle to be identified which moves through the first weight detection zone is collected in real time; when the weight of the cargo is the weight of the casting object cast by the vehicle to be identified, as shown in the diagram (b) in fig. 3, a gravity sensor, a weight sensor or a pressure sensor is arranged on the left and right sides of the road between the target area and the truck weighing area at intervals of a preset length to form a second weight detection area, and when the vehicle to be identified drives from the target area to the truck weighing area, the weight of the casting object falling into the second weight detection area is detected in real time to obtain the weight of the casting object cast by the vehicle to be identified.

In an alternative embodiment, in order to reduce the number of deployed sensors and reduce the hardware cost, a target area may be preset at a preset distance in front of a truck weighing area, a gravity sensor, a weight sensor or a pressure sensor may be deployed in a road between the target area and the truck weighing area, a camera, an image sensor and/or a radar may be deployed in the truck weighing area, when a vehicle is detected to drive into the target area in the target lane, the vehicle may be set as a vehicle to be identified, prompt information may be sent to prompt the vehicle to be identified to drive into the truck weighing area, and the weight of goods in the process of driving the vehicle from the target area to the truck weighing area may be acquired through the deployed gravity sensor, the weight sensor or the pressure sensor, and the contour information, the driving track of the vehicle to be identified in the process of driving the target area to the truck weighing area may be acquired through the deployed camera, the image sensor and/or the radar, or the image sequence of the vehicle to be identified in the preset duration may be acquired, when the vehicle to be detected to drive into the target area, the vehicle to be identified, or the image of the vehicle to be identified may be acquired through the camera, the image sensor after the vehicle to be identified.

220, carrying out sprinkling event identification according to the data to be identified, and determining whether the sprinkling event occurs to the vehicle to be identified.

In some alternative embodiments, the data to be identified may be compared with pre-stored reference identification data, and whether the vehicle to be identified has a throwing event may be determined according to the comparison result. By taking the data to be identified as the truck weighing area image as an example, the truck weighing area image can be compared with a pre-stored reference truck weighing area image, if the truck weighing area image is consistent with the pre-stored reference truck weighing area image, the condition that the truck to be identified does not have a throwing event is determined, and if the truck weighing area image is inconsistent with the pre-stored reference truck weighing area image, the condition that the truck to be identified has a throwing event is determined. The reference truck weighing area image can be a pre-collected image without throwing objects in the truck weighing area, and can also be an image of the truck weighing area collected after the former vehicle to be identified drives out of the truck weighing area.

In some optional embodiments, the risk of throwing the data to be identified can be identified through a preset identification model, and whether the vehicle to be identified has a throwing event or not is determined. Alternatively, the recognition model may be a machine learning model, a probabilistic model, or a neural network model.

In some optional embodiments, the risk of shedding is identified according to the data to be identified, so as to obtain the risk degree of shedding of the vehicle to be identified, and whether the vehicle to be identified has a shedding event is determined according to the risk degree of shedding of the vehicle to be identified.

The vehicle to be identified is characterized by whether a vehicle to be identified is easy to generate a throwing event, and the higher the throwing risk degree is, the easier the vehicle to be identified is to generate the throwing event. For example, taking the example that the shedding risk degree includes low risk, medium risk and high risk, when the shedding risk degree of the vehicle to be identified is low risk, the vehicle to be identified is indicated to have no shedding event and is not easy to generate the shedding event, the vehicle to be identified is determined to have no shedding event and is easy to generate the shedding event when the shedding risk degree of the vehicle to be identified is medium risk, the vehicle to be identified is determined to have the shedding event, when the shedding risk degree of the vehicle to be identified is high risk, the vehicle to be identified is indicated to have the shedding event, and the vehicle to be identified is determined to have the shedding event. It should be noted that, the foregoing risk level is described as an example, a specific type of the risk level may be determined according to an actual application scenario, for example, the risk level may be represented by a numerical value, where the greater the numerical value, the greater the risk level, and the easier the vehicle to be identified is to generate a casting event.

In an alternative embodiment, the risk of throwing the vehicle to be identified can be obtained by carrying out risk of throwing identification on the data to be identified through a preset identification model. Alternatively, the recognition model may be a machine learning model, a probabilistic model, or a neural network model.

In an alternative embodiment, the data to be identified and preset data to be identified of each risk degree of throwing can be matched, and the risk degree of throwing of the vehicle to be identified is obtained according to the matching result. The matching result represents the similarity degree of the data to be identified and the preset data to be identified.

230, if the vehicle to be identified is subjected to a throwing event, outputting first prompt information to prompt the vehicle to be identified to drive into a preset safety area and rectifying and modifying the goods loaded by the vehicle to be identified.

Optionally, if the vehicle to be identified has a throwing event, outputting a first prompt message, determining a preset first driving route as a target driving route of the vehicle to be identified, and prompting the vehicle to be identified to drive into a preset safety area through the target driving route to rectify and change the goods loaded by the vehicle to be identified.

As shown in fig. 1, the preset first travel route may be a preset first travel path that communicates a truck weighing zone and a preset safety zone; the preset safety zone is used for rectifying and modifying the vehicle with the risk degree larger than or equal to the preset risk degree threshold, wherein rectifying and modifying the cargoes loaded by the vehicle to be identified comprises, but is not limited to, reassembling the cargoes, reducing the assembled cargoes, adjusting the carriage and the like. Wherein, adjusting the carriage can be adding a covering cloth.

In this embodiment, in order to reduce the number of times of a throwing event occurring on a highway, and ensure the driving safety of the highway, when the throwing event occurs on the vehicle to be identified, a first prompting message is output, the vehicle to be identified is refused to enter the highway through the entrance of the highway toll gate, and a preset first driving route is determined as a target driving route of the vehicle to be identified, so as to prompt the vehicle to be identified to drive into a preset safety area through the target driving route, and modify the goods loaded by the vehicle to be identified.

Optionally, the first prompt information may be output through voice, or the first prompt information may be output through a display screen. The first prompt information may be alarm information.

Optionally, after the goods loaded by the vehicle to be identified are rectified, the rectified vehicle is restarted to enter the truck weighing area, the vehicle management system at the expressway entrance executes the step 210-220 to identify the rectified vehicle with the shedding event, determine whether the rectified vehicle has the shedding event, output second prompt information when the rectified vehicle does not have the shedding event, determine the preset second driving route as the target driving route of the rectified vehicle, prompt the rectified vehicle to drive into the expressway toll gate entrance, output first prompt information when the rectified vehicle has the shedding event, prompt the rectified vehicle to drive into the preset safety area for rectification again.

Optionally, after the goods loaded by the vehicle to be identified are rectified, the rectified vehicle is restarted to enter the truck weighing area, the vehicle management system of the expressway entrance executes step 210 to obtain new data to be identified, and according to the new data to be identified, the rectified vehicle is subjected to risk dumping identification to obtain the risk dumping degree of the rectified vehicle, when the risk dumping degree of the rectified vehicle is smaller than a preset risk degree threshold value, a first prompt message is output, a preset second driving route is determined as a target driving route of the rectified vehicle, so that the rectified vehicle is prompted to drive into the expressway toll station entrance, when the risk dumping degree of the rectified vehicle is larger than or equal to the preset risk degree threshold value, the rectified vehicle is refused to drive into the expressway toll station entrance, and the first prompt message is output, so that the rectified vehicle is prompted to drive into a preset safety area for being rectified again.

And 240, outputting second prompt information to prompt the vehicle to be identified to drive into the entrance of the expressway toll gate if the vehicle to be identified does not have a throwing event.

Optionally, if the vehicle to be identified does not have a throwing event, outputting a second prompt message, and determining the preset second driving route as the target driving route of the rectified vehicle, so as to prompt the rectified vehicle to drive into the entrance of the highway toll station.

As shown in fig. 1, the preset second driving route may be a preset second driving path that communicates the truck weighing zone and a preset safety zone.

According to the vehicle management method for the expressway entrance, provided by the embodiment of the invention, the truck weighing area is arranged at the preset distance from the expressway toll gate entrance, so that vehicles which are prevented from being thrown into the expressway can be timely found, and then the throwing objects on the driving road surface of the expressway are reduced, and the driving safety of the expressway is ensured; and the sprinkling event identification is carried out in a truck weighing area which is a preset distance away from the entrance of the expressway toll station, so that the probability of causing traffic jam can be reduced.

In an alternative embodiment, taking the data to be identified as the vehicle image sequence as an example for explanation, the vehicle to be identified can stay in the truck weighing zone for a preset time period through the vehicle image sequence to detect the sprinkled object, so as to determine whether the vehicle to be identified has a sprinkling event.

Optionally, the vehicle image sequence includes an image sequence in which the vehicle to be identified stops in the truck weighing zone for a preset period of time.

Optionally, when the vehicle to be identified is detected to drive into the truck weighing area, a video image of the vehicle to be identified in the truck weighing area can be acquired through a camera or an image sensor, so as to obtain a vehicle image sequence. Specifically, the acquisition method of the vehicle image sequence includes steps a1 to a2:

Step a1, determining whether the vehicle to be identified enters a truck weighing zone.

In some alternative embodiments, it may be determined whether the vehicle to be identified is driving into the wagon weighing zone by a weight detection data sequence acquired by a sensing unit in the wagon weighing zone. Wherein the sensing unit may be a weight sensor or a pressure sensor. The weight detection data sequence can be obtained by periodically detecting weight data in the weighing area of the truck by the sensing unit in a preset unit time length, and comprises weight data corresponding to each detection time in the preset unit time length.

Specifically, the method for determining whether a vehicle to be identified is driven into a truck weighing zone based on a weight detection data sequence comprises the following steps:

(1) And acquiring a weight detection data sequence of the sensing unit in the truck weighing area within a preset unit time.

(2) And determining the weight change trend according to the weight detection data sequence.

Wherein the weight change trend comprises weight increment and weight decrement.

(3) And if the weight change trend is that the weight is increased, determining that the vehicle to be identified is driven into the truck weighing area.

(4) If the weight change trend is that the weight is decreased, and the limit value of the weight detection data sequence is larger than or equal to a preset limit value threshold value, determining that the vehicle to be identified exits from the truck weighing zone.

Alternatively, the weight data of each detection time in the weight detection data sequence may be subjected to differential processing to obtain a weight change sequence of the weight detection data sequence, and the weight change trend may be determined according to the weight change sequence. Alternatively, the weight change sequence of the weight detection data sequence may be obtained by calculating a weight difference between the weight data of each detection time in the weight detection data sequence and the weight data of the previous detection time adjacent to the detection time.

Optionally, determining the weight change trend according to the weight change sequence includes: if all the values in the weight change sequence are equal to 0, indicating that the weight in the truck weighing area is not changed in unit time length, determining that the weight change trend is unchanged; if a plurality of continuous numerical values larger than 0 exist in the weight change sequence, and target weight data larger than preset weight data exist in the weight data of each detection time in the weight detection data sequence, namely that the weight in the truck weighing area is increasing in unit time length, determining that the weight change trend is increasing in weight; if a plurality of values smaller than 0 exist in the weight change sequence continuously, and the time period of the values smaller than 0 in the weight change sequence is longer than the preset time, which means that the weight in the truck weighing zone is decreasing in unit time, the weight change trend is determined to be decreasing in weight.

In other alternative embodiments, whether the vehicle to be identified is driven into the truck weighing zone may be determined based on depth information of the vehicle in successive frame images of the truck weighing zone within a preset unit duration. Specifically, the method for determining whether to drive in a vehicle to be identified in a truck weighing zone based on depth information comprises the following steps:

(1) And acquiring continuous frame images of the truck weighing area within a preset unit time length.

(2) And carrying out vehicle detection on each frame of images in the continuous frame of images to obtain a vehicle detection result of each frame of images. Wherein the vehicle detection characterizes whether a vehicle to be identified is detected in each frame of image.

(3) And determining a target frame image of the vehicle to be identified in the continuous frame images according to the vehicle detection result of each frame image, and extracting an image sequence of the vehicle to be identified from the continuous frame images by taking the target frame image as a starting image.

(4) And determining depth information of the vehicle to be identified in each image in the image sequence to obtain a depth information sequence of the vehicle to be identified. The depth information refers to the distance between the points corresponding to the position information in the image and the camera shooting the image when the points are mapped into the real truck weighing zone.

(5) And obtaining the depth information change trend of the vehicle to be identified according to the depth information sequence.

The depth information change trend is used for representing the change condition of the distance between the vehicle to be identified and the camera. The depth information change trend of the vehicle to be identified comprises decreasing of the corresponding value of the depth information of the vehicle to be identified, increasing of the corresponding value of the depth information of the vehicle to be identified and unchanged of the corresponding value of the depth information of the vehicle to be identified in a time period corresponding to a plurality of continuous image sequences. Optionally, when the change trend of the depth information decreases, that is, the distance between the vehicle to be identified and the camera gradually decreases, it is indicated that the vehicle to be identified is approaching the camera, and it is determined that the vehicle to be identified is driving into the truck weighing area.

(6) If the change trend of the depth information is continuously decreasing, determining that the vehicle to be identified is driven into the truck weighing area.

(7) If the change trend of the depth information is continuously increased, and the depth information of the vehicle to be identified in each image has target depth information larger than a preset value, determining that the vehicle to be identified exits from the truck weighing area.

The preset value may be a length or a width of the truck weighing zone, and the preset value may also be a maximum distance between the truck weighing zone and a camera disposed in the truck weighing zone.

Optionally, vehicle detection may be performed on each of the continuous frame images to determine whether a vehicle is present in the continuous frame images; if no vehicle exists in the continuous frame images, indicating that no vehicle exists in the truck weighing area in the corresponding time period of the continuous frame images, acquiring the continuous frame images of the next preset unit duration; and if the vehicle exists in the continuous frame images, extracting an image sequence with the vehicle to be identified from the continuous frame images by taking the target frame image as a starting image for determining the target frame image with the vehicle to be identified in the continuous frame images for the first time.

In some embodiments, vehicle detection may be performed on each of the successive frame images by image differencing each of the successive frame images. Specifically, an image when no vehicle exists in a truck weighing area is collected and set as a reference image, and each frame of image in continuous frame images is subjected to difference with the reference image frame by frame to obtain a difference image; acquiring an area with a pixel value larger than a preset pixel value in the differential image, and if the average value of the pixel values of the area is larger than the preset average value, indicating that the frame image is different from the reference image, determining whether the average value of the pixel values of the area with the pixel value larger than the preset pixel value in the differential image of the subsequent frame image in the frame image is larger than the preset average value; if the average value of the pixel values of the areas with the pixel values larger than the preset pixel values in the differential image of the subsequent frame image is larger than the preset average value, determining that a vehicle exists in the continuous frame image; if the average value of the pixel values of the region is smaller than or equal to the preset average value, or if the pixel value in the differential image does not exist in the region larger than the preset pixel value, indicating that the frame image is the same as the reference image, determining whether the differential image of the subsequent frame image in the continuous frame image exists in the region larger than the preset pixel value; and if no area larger than the preset pixel value exists in the differential image of the subsequent frame image in the continuous frame image, determining that no vehicle exists in the continuous frame image.

In other embodiments, each of the successive frame images may be vehicle detected by determining whether a car is present in each of the successive frame images, such as by a trained car detection model. The car detection model can be a machine learning model, such as dictionary learning and logistic regression models, and can also be a detection model based on a neural network. Such as YOLO-based detection models, SSD-based detection models, RCNN-based detection models.

In some embodiments, depth detection may be performed on each image in the sequence of images to determine depth information of the vehicle to be identified in each image in the sequence of images.

Optionally, target detection may be performed on each image in the image sequence, an image area of the vehicle to be identified in each image in the image sequence is determined, and depth estimation is performed on the image area of the vehicle to be identified in each image in the image sequence, so as to obtain depth information of the vehicle to be identified in each image in the image sequence. The depth estimation network may be used to estimate the depth of the image region of the vehicle to be identified in each image in the image sequence, and the depth estimation network may be a residual network or a network based on an attention mechanism.

In other embodiments, the target detection may be performed on each image in the image sequence to obtain a boundary frame coordinate of the vehicle to be identified in each image, a midpoint coordinate of the boundary frame of the vehicle to be identified is calculated according to the boundary frame coordinate of the vehicle to be identified, the midpoint coordinate of the boundary frame of the vehicle to be identified is set as the position information of each image in the image sequence of the vehicle to be identified in the truck weighing area, and the preset position depth relation data is queried according to the position information to obtain the depth information of the vehicle to be identified in each image in the image sequence.

The position depth relation data comprise each piece of position information in the truck weighing zone and depth information corresponding to the position information. The boundary frame coordinates of the vehicle to be identified may be boundary frame coordinates of a cabin frame of the vehicle to be identified, such as corner coordinates of a cabin boundary, center point coordinates of convenience of a cabin, midpoint coordinates of an upper boundary of the cabin boundary, or midpoint coordinates of a lower boundary of the cabin boundary.

Optionally, when no vehicle is in the truck weighing area, an image of the truck weighing area is acquired, depth estimation is performed on the image of the truck weighing area through a trained depth estimation network, depth information of each pixel point in the image of the truck weighing area is obtained, position information of each pixel point in the image of the truck weighing area is obtained, and the position information of each pixel point in the image of the truck weighing area is stored in a correlated manner with the depth information corresponding to the pixel point, so that position depth relation data is obtained. Alternatively, the depth estimation network may be a residual network or a network based on an attention mechanism.

In some optional embodiments, after determining the depth information of the vehicle to be identified in each image in the image sequence, each image in the image sequence may be ranked according to the time sequence of each image in the image sequence, so as to obtain a ranked image sequence and the depth information of the vehicle to be identified in each image in the ranked image sequence, and the depth information sequence of the vehicle to be identified is obtained according to the depth information of the vehicle to be identified in each image in the ranked image sequence.

Optionally, the depth information of the target vehicle in the sorted transportation images may be differenced with a preset distance threshold to obtain a difference sequence, and the difference sequence is set as the depth information sequence of the vehicle to be identified. It should be noted that, in the embodiment of the present application, the preset distance threshold is not specifically limited, and may be set according to an actual application scenario, for example, the preset distance threshold may be 1 meter.

Alternatively, the depth information of the vehicle to be identified in each image in the ordered image sequence may be set as the depth information sequence of the vehicle to be identified.

In some optional embodiments, obtaining the depth information change trend of the vehicle to be identified according to the depth information sequence includes: and carrying out forward difference on the depth information sequence of the vehicle to be identified to obtain a difference result, and obtaining the depth information change trend of the vehicle to be identified according to the difference result. For example, when the difference result is smaller than 0, it is indicated that the difference corresponding to the next time is smaller than the difference corresponding to the previous time, and the variation trend of the depth information of the vehicle to be identified is obtained to be decreasing; when the difference result is larger than 0, the difference corresponding to the next moment is larger than the difference corresponding to the previous moment, and the change trend of the depth information of the vehicle to be identified is obtained to be incremental.

In some optional embodiments, to further determine whether the vehicle to be identified drives into the truck weighing area, after the depth information change trend of the vehicle to be identified, if the depth information change trend is continuously decreasing and at least one target depth information smaller than or equal to the preset depth information exists in the depth information sequence of the vehicle to be identified, determining that the vehicle to be identified drives into the truck weighing area, and if the depth information change trend is continuously increasing and the target depth information larger than the preset value exists in the depth information sequence of the vehicle to be identified, determining that the vehicle to be identified drives out of the truck weighing area.

And a2, if the vehicle to be identified is driven into the truck weighing area, acquiring video images of the vehicle to be identified, and obtaining a vehicle image sequence.

In some alternative embodiments, when the vehicle to be identified is driven into the truck weighing zone, the vehicle management system of the expressway entrance is started, video images of the vehicle to be identified which stays in the truck weighing zone are collected, and when the vehicle to be identified is detected to output the truck weighing zone through the step a1, the video images of the vehicle to be identified, which are collected between the time when the vehicle to be identified is driven into the truck weighing zone and the time when the vehicle to be identified outputs the truck weighing zone, are set as a vehicle image sequence.

In some alternative embodiments, when the vehicle to be identified is driven into the truck weighing zone, the vehicle management system of the expressway entrance is started, video images of the vehicle to be identified staying in the truck weighing zone are collected in real time, and the video images of the vehicle to be identified collected in real time are set as a vehicle image sequence.

Optionally, in some embodiments, the target detection may be performed on each vehicle image in the sequence of vehicle images to obtain a casting detection result, and whether the vehicle to be identified has a casting event is determined according to the casting detection result. Wherein the casting detection results are indicative of whether casting is present in each vehicle image. Specifically, the vehicle image sequence-based throwing event recognition method comprises the steps of b 1-b 3:

and b1, carrying out target detection on each vehicle image in the vehicle image sequence, and determining whether a casting object exists in each vehicle image.

And b2, if no casting object exists in each vehicle image in the vehicle image sequence, determining that the vehicle to be identified does not have a casting event.

And b3, if at least one target vehicle image exists in the vehicle image sequence, determining that the vehicle to be identified has a throwing event.

Optionally, in some optional embodiments, each vehicle image may be matched with a preset reference image, and the projectile detection result is determined according to the matching result. The preset reference image may be an image of the pre-collected truck weighing area where no casting exists. If the vehicle image is not matched with the preset reference image, determining that the casting object detection result of the vehicle image is that the casting object exists in the vehicle image; if the vehicle image is matched with the preset reference image, determining that the casting object detection result of the vehicle image is that the casting object does not exist in the vehicle image.

Optionally, each vehicle image and a preset reference image may be subjected to differential processing to obtain a differential image, and a matching result is obtained according to the pixel value of the pixel point in the differential image. For example, when the pixel values of the pixels of the differential image are all preset pixel values, it is determined that the vehicle image is matched with the preset reference image, and when target pixel points with the pixel values not being the preset pixel values exist in the pixels of the differential image and the number of the pixels of the target pixel points is larger than the preset number of the pixels, it is determined that the vehicle image is not matched with the preset reference image.

In some optional embodiments, for each vehicle image in the vehicle image sequence, the vehicle image and a previous vehicle image adjacent to the vehicle image may be differentiated to obtain a vehicle differential image of the vehicle, and whether the throwing object exists in each vehicle image is determined according to the pixel value of the pixel point in each vehicle differential image. For example, when the pixel values of the pixel points of each vehicle differential image are all preset pixel values, it is determined that no casting is present in each vehicle image in the vehicle image sequence, and when target vehicle differential images with the pixel values of the pixel points not being the preset pixel values and with the pixel number of the target pixel points being greater than the preset pixel number exist in each pixel point in each vehicle differential image, it is determined that at least one target vehicle image is present in the vehicle image sequence.

In some optional embodiments, feature extraction may be performed on each vehicle image to obtain an image feature of each vehicle image, and target detection may be performed according to the image feature to determine that a casting detection result is obtained. Including but not limited to image texture features, image contrast features, image edge features, image semantic features, etc.

Alternatively, feature extraction can be performed on each vehicle image by an edge detection operator, a texture feature extraction method and a gray level detection method, so as to obtain the image feature of each vehicle image. And extracting the characteristics of each vehicle image through a preset characteristic extraction network to obtain the image characteristics of each vehicle image.

By way of example, taking a case that feature extraction is performed on each vehicle image through a preset feature extraction network as shown in fig. 4, fig. 4 is a schematic structural diagram of the feature extraction network provided by the embodiment of the present invention, where the feature extraction network includes a convolution layer and a feature extraction layer, the convolution layer of the preset feature extraction network performs convolution processing on each vehicle image to obtain an initial feature image of each vehicle image, the feature extraction layer performs feature extraction on the initial feature image to obtain an image feature of each vehicle image, and the image feature of each vehicle image is input into a detection network to perform target detection to determine that a sprinkle detection result is obtained.

The feature extraction layer includes, as shown in fig. 4, a residual unit, a first residual group, a second residual group, a third residual group, a first deconvolution unit, a second deconvolution unit, a first fusion unit, and a second fusion unit. Specifically, a residual unit of the feature extraction layer carries out residual processing on the initial feature map, and the processed initial feature map is input into a first residual group; the first residual group carries out residual processing and receptive field enhancement on the input processed initial feature image to obtain a first residual feature image, and the first residual feature image is respectively input into a first fusion unit and a second residual group; the second residual group carries out residual processing and receptive field enhancement on the first residual feature map to obtain a second residual feature map, and the second residual feature map is respectively input into a first deconvolution unit, a second fusion unit and a third residual group; performing residual processing and receptive field enhancement on the second residual feature map by a third residual group to obtain a third residual feature map, and respectively inputting the third residual feature map into a second deconvolution unit; the first deconvolution unit carries out deconvolution on the second residual characteristic map and then inputs the second residual characteristic map into the first fusion unit, and the first fusion unit fuses the first residual characteristic map and the deconvolution-processed second residual characteristic map to obtain a first fusion characteristic map; the second deconvolution carries out deconvolution on the third residual error map and then inputs the third residual error map into a second fusion unit, and the second fusion unit fuses the second fusion feature map and the deconvolution-processed third residual error feature map to obtain a second fusion feature map; and setting the first fusion feature map, the second fusion feature map and the third residual feature map as image features of each vehicle image.

The first residual group, the second residual group and the third residual group are similar in structure and comprise a residual sub-network and a receptive field enhancement unit which are connected in series. Wherein the receptive field enhancement unit is used for receptive field enhancement of image features of which residuals are output from the network, and optionally, the receptive field enhancement unit comprises a first convolution kernel of 3*3 and a second convolution kernel of 1*1. It should be noted that, in the embodiment of the present invention, the network structure of the residual sub-network corresponding to each of the first residual group, the second residual group and the third residual group is not specifically limited, and the network structure of the residual sub-network corresponding to each of the first residual group, the second residual group and the third residual group may be set according to an actual application scenario.

Optionally, regression processing may be performed on the image features to obtain a probability of the presence of the casting object in each vehicle image, and the probability of the presence of the casting object in each vehicle image is compared with a preset probability threshold, and if the probability of the presence of the casting object in each vehicle image is greater than or equal to the preset probability threshold, it is determined that the casting object exists as the detection result of the casting object, and if the probability of the presence of the casting object in each vehicle image is less than the preset probability threshold, it is determined that the casting object does not exist as the detection result of the casting object. In some implementations, the image features may be regressed by a machine-learning based model, such as by a regression model.

Optionally, the image features may be input to a prediction network to perform the detecting of the sprinkle, and the detecting result of the sprinkle is determined. By way of example, the predictive network may be a fully connected network.

In some optional embodiments, taking the data to be identified as the image of the truck weighing area as an example for explanation, when the condition that the vehicle to be identified drives out of the truck weighing area is determined, the image of the truck weighing area can be acquired, the image of the truck weighing area is obtained, and whether the vehicle to be identified has a throwing event or not is determined according to the image of the truck weighing area. Specifically, the method for determining whether the vehicle to be identified has a throwing event or not based on the truck weighing zone image comprises the following steps of c 1-c 5:

step c1, determining whether the vehicle to be identified is driven out of the weighing area of the truck.

In some embodiments, whether the vehicle to be identified is driven out of the truck weighing bay may be determined according to the determination method of whether the vehicle to be identified is driven in the truck weighing bay based on the depth information or according to the determination method of whether the vehicle to be identified is driven in the truck weighing bay based on the weight detection data sequence.

And c2, if the vehicle to be identified exits the truck weighing area, acquiring an image of the truck weighing area to obtain an image of the truck weighing area.

And c3, comparing the truck weighing area image with the historical truck weighing area image. The historical truck weighing area image is an image of the truck weighing area acquired after the previous to-be-identified truck exits the truck weighing area.

And c4, if the truck weighing area image is matched with the historical truck weighing area image, determining that the vehicle to be identified is not subjected to a throwing event.

And c5, if the truck weighing area image is not matched with the historical truck weighing area image, determining that a vehicle to be identified is subjected to a throwing event.

In some alternative embodiments, it may be determined from the difference image between the truck weighing zone image and the historical truck weighing zone image whether the truck weighing zone image matches the historical truck weighing zone image. For example, if the pixel value of each pixel point in the difference image is consistent with the preset pixel value, determining that the truck weighing area image is matched with the historical truck weighing area image, and if the target pixel point, of which the pixel value is inconsistent with the preset pixel value, exists in the difference image, determining that the truck weighing area image is not matched with the historical truck weighing area image.

In some alternative embodiments, the feature extraction network shown in fig. 4 may perform feature extraction on the truck weighing zone image and the historical truck weighing zone image respectively to obtain a first feature image of the truck weighing zone image and obtain a second feature image of the historical truck weighing zone image, if the first feature image is consistent with the second feature image, it is determined that the truck weighing zone image is matched with the historical truck weighing zone image, and if the first feature image is inconsistent with the second feature image, it is determined that the truck weighing zone image is not matched with the historical truck weighing zone image.

In some alternative embodiments, after determining whether the vehicle to be identified has a dumping event according to the truck weighing area image and the historical truck weighing area image, a new historical truck weighing area image may be set in place of the historical truck weighing area image in the currently acquired truck weighing area image, and when detecting that a new vehicle to be identified exits from the truck weighing area, the new truck weighing area image is acquired, and the steps c3 to c5 are executed according to the new truck weighing area image and the new historical truck weighing area image to determine whether the vehicle is identified as having a dumping event.

In some optional embodiments, to further ensure the driving safety of the expressway, when it is determined that the vehicle to be identified does not have a casting event, the probability of the casting event of the vehicle to be identified may be predicted according to the vehicle data of the vehicle to be identified, if the probability of the casting event of the vehicle to be identified is greater than or equal to a preset probability threshold, a first prompt message is output to prompt the vehicle to be identified to drive into a preset safety area to modify the cargo loaded by the vehicle to be identified, and if the probability of the casting event of the vehicle to be identified is less than the preset probability threshold, a second prompt message is output to prompt the vehicle to be identified to drive into the entrance of the expressway toll station.

The vehicle data of the vehicle to be identified includes, but is not limited to, cargo weight and carriage profile information of the vehicle to be identified. The cargo weight can be the weight of the vehicle body of the vehicle to be identified, or the weight of a throwing object thrown by the vehicle to be identified. The car profile information includes, but is not limited to, car length, car height, car volume, car type, including, but not limited to, closed car, semi-closed car, etc.

Alternatively, the probability of a vehicle to be identified being subjected to a dumping event may be predicted based on one or more of the cargo weight, cargo type, and car profile information in the vehicle data.

Taking the cargo weight and the compartment contour information in the vehicle data as an example, the probability of the vehicle to be identified that a throwing event occurs can be predicted according to the cargo weight and the cargo type of the vehicle to be identified and the mapping relation between the pre-stored vehicle data and the probability. The mapping relation between the vehicle data and the probability can be a mapping data table between the vehicle data and the probability, wherein the mapping data table comprises a plurality of preset cargo types and preset probabilities of the occurrence of a throwing event of each preset cargo type under different preset cargo weights, and the mapping data table between the vehicle data and the probability is inquired according to the cargo weight and the cargo types of the vehicle to be identified to obtain the probability of the occurrence of the throwing event of the cargo types of the vehicle to be identified under the cargo weight; the mapping relation between the vehicle data and the probability may also be a mapping function between the vehicle data and the probability, such as a linear function, a mixed function, an exponential function, a power function, etc., and the cargo type of the vehicle to be identified is subjected to numerical conversion to obtain cargo type parameters of the vehicle to be identified, and the cargo weight and cargo type parameters of the vehicle to be identified are input into the mapping function between the vehicle data and the probability to perform probability prediction to obtain the probability of the vehicle to be identified having a throwing event. It should be noted that, the specific form of the mapping relationship between the vehicle data and the probability is not limited in the embodiment of the present invention.

Taking the cargo weight and the carriage profile information in the vehicle data as an example, probability prediction can also be performed according to the cargo weight and the carriage profile information of the vehicle to be identified and the mapping relation between the pre-stored vehicle data and the probability, so as to obtain the probability of the vehicle to be identified from the throwing event. The cargo weight and the carriage profile information of the vehicle to be identified can be input into a preset casting event prediction model, and the probability of casting event of the vehicle to be identified is obtained. The shedding event prediction model can be based on a machine learning prediction model, a probability model or a neural network prediction model.

According to the vehicle management method for the expressway entrance, provided by the embodiment of the invention, the truck weighing area is arranged at the preset distance from the expressway toll gate entrance, so that vehicles which are easy to induce a throwing event can be timely found and excluded from entering the expressway, and then throwing objects on the driving road surface of the expressway are reduced, and the driving safety of the expressway is ensured; and the throwing event identification is carried out on the data to be identified of the truck weighing area which is at a preset distance from the entrance of the expressway toll station, so that the probability of causing traffic jam can be reduced.

In order to better implement the vehicle management method for the expressway entrance provided by the embodiment of the invention, on the basis of the vehicle management method embodiment for the expressway entrance, a vehicle management device for the expressway entrance is provided, as shown in fig. 5, fig. 5 is a schematic structural diagram of the vehicle management device for the expressway entrance provided by the embodiment of the invention, and the vehicle management device for the expressway entrance includes:

the acquisition module is used for acquiring the data to be identified of the truck weighing area after the vehicle to be identified stays in the truck weighing area for a preset time period; the truck weighing area is spaced from the entrance of the expressway toll station by a preset distance;

the recognition module is used for recognizing the throwing event according to the data to be recognized and determining whether the vehicle to be recognized has the throwing event or not;

the first processing module is used for outputting first prompt information to prompt the vehicle to be identified to drive into a preset safety area and to rectify and change the goods loaded by the vehicle to be identified if the vehicle to be identified has a throwing event;

In an alternative embodiment, the data to be identified comprises a sequence of vehicle images comprising a sequence of images of the vehicle to be identified in a predetermined time period of stay in the truck weighing zone, the identification module being configured to:

performing target detection on each vehicle image in the vehicle image sequence, and determining whether a casting object exists in each vehicle image;

if no throwing object exists in each vehicle image in the vehicle image sequence, determining that a throwing event does not occur in the vehicle to be identified;

and if the at least one target vehicle image exists in the vehicle image sequence, determining that the vehicle to be identified has a throwing event.

In an alternative embodiment, the acquisition module is configured to:

determining whether a vehicle to be identified enters a truck weighing zone;

and if the vehicle to be identified is driven into the truck weighing area, acquiring video images of the vehicle to be identified, and obtaining a vehicle image sequence.

In an alternative embodiment, the acquisition module is configured to:

acquiring a weight detection data sequence of a sensing unit in a truck weighing area within a preset unit time length;

determining a weight change trend according to the weight detection data sequence; the weight change trend includes increasing weight and decreasing weight;

If the weight change trend is gradually increased, determining that the vehicle to be identified is driven into the truck weighing area;

if the weight change trend is that the weight is decreased, and the limit value of the weight detection data sequence is larger than or equal to a preset limit value threshold value, determining that the vehicle to be identified exits from the truck weighing zone.

In an alternative embodiment, the acquisition module is configured to:

acquiring continuous frame images of a truck weighing area in a preset unit time length;

carrying out vehicle detection on each frame of images in the continuous frame images to obtain a vehicle detection result of each frame of images; the vehicle detection represents whether a vehicle to be identified is detected in each frame of image;

according to the vehicle detection result of each frame of image, determining a target frame image of the vehicle to be identified in the continuous frame images for the first time, and extracting an image sequence with the vehicle to be identified from the continuous frame images by taking the target frame image as a starting image;

determining depth information of the vehicle to be identified in each image in the image sequence to obtain a depth information sequence of the vehicle to be identified;

obtaining the depth information change trend of the vehicle to be identified according to the depth information sequence;

if the change trend of the depth information is continuously decreasing, determining that the vehicle to be identified is driven into the truck weighing area;

If the change trend of the depth information is continuously increased, and the depth information of the vehicle to be identified in each image has target depth information larger than a preset value, determining that the vehicle to be identified exits from the truck weighing area.

In an alternative embodiment, the data to be identified includes a truck weighing area image, where the truck weighing area image is an image of a truck weighing area acquired after the vehicle to be identified exits the truck weighing area, and the identification module is configured to:

determining whether the vehicle to be identified drives out of the weighing area of the truck;

if the vehicle to be identified drives out of the truck weighing area, acquiring an image of the truck weighing area to obtain an image of the truck weighing area;

comparing the truck weighing area image with the historical truck weighing area image; the historical truck weighing area image is an image of the truck weighing area acquired after the previous to-be-identified truck exits the truck weighing area;

if the truck weighing area image is matched with the historical truck weighing area image, determining that a vehicle to be identified does not have a throwing event;

and if the truck weighing area image is not matched with the historical truck weighing area image, determining that the vehicle to be identified is subjected to a throwing event.

In an alternative embodiment, the identification module is configured to:

and setting a new historical truck weighing zone image for the truck weighing zone image to replace the historical truck weighing zone image.

The embodiment of the invention also provides a vehicle management device for an expressway entrance, as shown in fig. 6, which shows a schematic structural diagram of the vehicle management device for an expressway entrance according to the embodiment of the invention, specifically:

the highway entry vehicle management device may include one or more processors 401 of a processing core, one or more memories 402 of a computer readable storage medium, a power supply 403, and an input unit 404. It will be appreciated by those skilled in the art that the configuration of the vehicle management device for an expressway portal shown in fig. 6 does not constitute a limitation of the vehicle management device for an expressway portal, and may include more or less components than those illustrated, or may combine certain components, or a different arrangement of components. Wherein:

the processor 401 is a control center of the vehicle management apparatus of the expressway portal, connects respective parts of the vehicle management apparatus of the entire expressway portal with various interfaces and lines, and performs various functions and processes of the vehicle management apparatus of the expressway portal by running or executing software programs and/or modules stored in the memory 402 and calling data stored in the memory 402, thereby performing overall monitoring of the vehicle management apparatus of the expressway portal. Optionally, processor 401 may include one or more processing cores; preferably, the processor 401 may integrate an application processor and a modem processor, wherein the application processor mainly processes an operating system, a user interface, an application program, etc., and the modem processor mainly processes wireless communication. It will be appreciated that the modem processor described above may not be integrated into the processor 401.

The memory 402 may be used to store software programs and modules, and the processor 401 executes various functional applications and data processing by executing the software programs and modules stored in the memory 402. The memory 402 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program (such as a sound playing function, an image playing function, etc.) required for at least one function, and the like; the storage data area may store data created according to the use of the vehicle management apparatus of the highway entrance, and the like. In addition, memory 402 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 402 may also include a memory controller to provide the processor 401 with access to the memory 402.

The vehicle management apparatus for the highway entrance further includes a power supply 403 for supplying power to each component, and preferably, the power supply 403 may be logically connected to the processor 401 through a power management system, so that functions of managing charge, discharge, power consumption management, etc. are implemented through the power management system. The power supply 403 may also include one or more of any of a direct current or alternating current power supply, a recharging system, a power failure detection circuit, a power converter or inverter, a power status indicator, and the like.

The highway entrance vehicle management apparatus may further include an input unit 404, and the input unit 404 may be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

Although not shown, the vehicle management apparatus of the expressway entrance may further include a display unit or the like, which is not described herein. In particular, in this embodiment, the processor 401 in the vehicle management device of the highway entrance loads the executable files corresponding to the processes of one or more application programs into the memory 402 according to the following instructions, and the processor 401 executes the application programs stored in the memory 402, so as to implement various functions as follows:

if a vehicle to be identified is subjected to a throwing event, outputting first prompt information to prompt the vehicle to be identified to drive into a preset safety area and rectifying and modifying goods loaded on the vehicle to be identified;

And if the vehicle to be identified does not have a throwing event, outputting second prompting information to prompt the vehicle to be identified to drive into the entrance of the expressway toll gate.

Those of ordinary skill in the art will appreciate that all or a portion of the steps of the various methods of the above embodiments may be performed by instructions, or by instructions controlling associated hardware, which may be stored in a computer-readable storage medium and loaded and executed by a processor.

To this end, an embodiment of the present invention provides a storage medium having stored therein a plurality of instructions capable of being loaded by a processor to perform the steps of any one of the methods for managing vehicles at highway entrances provided in the embodiment of the present invention. For example, the instructions may perform the steps of:

The specific implementation of each operation above may be referred to the previous embodiments, and will not be described herein.

Wherein the storage medium may include: read Only Memory (ROM), random access Memory (RAM, random Access Memory), magnetic or optical disk, and the like.

The steps in any vehicle management method for an expressway entrance provided by the embodiment of the present invention may be executed by the instructions stored in the storage medium, so that the beneficial effects that any vehicle management method for an expressway entrance provided by the embodiment of the present invention may be achieved are detailed in the previous embodiments and are not described herein.

The foregoing has described in detail the methods, apparatus, devices and storage medium for vehicle management at highway entrances provided by the embodiments of the present invention, and specific examples have been applied herein to illustrate the principles and embodiments of the present invention, the above description of the embodiments being only for aiding in the understanding of the methods and core ideas of the present invention; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in light of the ideas of the present invention, the present description should not be construed as limiting the present invention.

Claims

1. A method of vehicle management for an expressway portal, the method comprising:

2. The method of vehicle management for highway entry according to claim 1, wherein said data to be identified comprises a sequence of vehicle images including a sequence of images of said vehicle to be identified in said truck weighing zone for a preset duration of stay;

the step of identifying the throwing event according to the data to be identified, and the step of determining whether the throwing event occurs to the vehicle to be identified comprises the following steps:

if no throwing object exists in each vehicle image in the vehicle image sequence, determining that the vehicle to be identified does not have a throwing event;

and if the object vehicle image exists in at least one target vehicle image in the vehicle image sequence, determining that the vehicle to be identified has a throwing event.

3. The method for vehicle management at highway entrance according to claim 2, wherein said method comprises, prior to said target detection of each of said sequence of vehicle images:

determining whether the vehicle to be identified enters the truck weighing zone;

and if the vehicle to be identified drives into the truck weighing area, acquiring video images of the vehicle to be identified to obtain a vehicle image sequence.

4. A method of vehicle management at a highway entrance according to claim 3 wherein said determining whether said vehicle to be identified is entering said truck weighing zone comprises:

Determining a weight change trend according to the weight detection data sequence; the weight change trend comprises weight increment and weight decrement;

and if the weight change trend is that the weight is decreased, and the limit value of the weight detection data sequence is larger than or equal to a preset limit value threshold, determining that the vehicle to be identified exits from the truck weighing zone.

5. A method of vehicle management at a highway entrance according to claim 3 wherein said determining whether said vehicle to be identified is entering said truck weighing zone comprises:

acquiring continuous frame images of the truck weighing area within a preset unit time length;

carrying out vehicle detection on each frame of images in the continuous frame of images to obtain a vehicle detection result of each frame of images; the vehicle detection characterizes whether a vehicle to be identified is detected in each frame of image;

according to the vehicle detection result of each frame of the images, determining a target frame image of the vehicle to be identified, which is detected for the first time in the continuous frame images, and extracting an image sequence of the vehicle to be identified from the continuous frame images by taking the target frame image as a starting image;

and if the change trend of the depth information is continuously increased, and the depth information of the vehicle to be identified in each image has target depth information larger than a preset value, determining that the vehicle to be identified exits from the truck weighing area.

6. The vehicle management method of an expressway entrance according to claim 1, wherein said data to be identified includes a truck weighing zone image, said truck weighing zone image being an image of said truck weighing zone acquired after said vehicle to be identified exits said truck weighing zone;

determining whether the vehicle to be identified exits the truck weighing zone;

if the vehicle to be identified exits the truck weighing area, acquiring an image of the truck weighing area to obtain an image of the truck weighing area;

Comparing the truck weighing area image with a historical truck weighing area image; the historical truck weighing area image is an image of the truck weighing area acquired after a previous vehicle to be identified exits the truck weighing area;

if the truck weighing area image is matched with the historical truck weighing area image, determining that the vehicle to be identified does not have a throwing event;

7. The method for managing vehicles at an entrance to an expressway according to claim 6, wherein said determining that said vehicle to be identified has not suffered a casting event or said determining that said vehicle to be identified has suffered a casting event comprises:

and setting a new historical truck weighing zone image for the truck weighing zone image so as to replace the historical truck weighing zone image.

8. A vehicle management apparatus for an expressway portal, the apparatus comprising:

9. A vehicle management apparatus for an expressway portal, comprising a memory and a processor; the memory stores an application program, and the processor is configured to execute the application program in the memory to perform the steps in the vehicle management method for an expressway entrance according to any one of claims 1 to 7.

10. A storage medium storing a plurality of instructions adapted to be loaded by a processor to perform the steps in the method of vehicle management for highway entry according to any one of claims 1 to 7.