KR102458136B1 - Bad load truck enforcement system - Google Patents

Abstract

A poorly loaded truck enforcement system is disclosed. The enforcement system for cracking down on a truck with bad loading according to an embodiment includes: a vehicle information acquisition unit configured to acquire rendering information of a vehicle generated using image information of a loading box of a vehicle photographed using a camera and sensor information; an input unit for inputting the acquired vehicle image information and the acquired vehicle rendering information into a learned learning model to crack down on a poorly loaded truck; and an illegal loading detection unit for detecting an illegally loaded vehicle from the vehicle's loading box image information and vehicle rendering information for cracking down on the poorly loaded truck input to the learned learning model.

Description

Bad Load Truck Enforcement System {BAD LOAD TRUCK ENFORCEMENT SYSTEM}

The description below relates to techniques for cracking down on badly loaded trucks.

In general, the Road Traffic Act applies to all types of vehicles traveling on the road. The Road Traffic Act is a law enacted to stipulate necessary matters regarding road traffic in order to secure safe and smooth traffic by preventing and removing all traffic hazards and obstacles that occur on the road. According to Article 39 of the Road Traffic Act, vehicles with poor solidarity that operate without securely fastening such as covering or tying the cover to prevent cargo from falling during operation, and vehicles without a cover that transports sand, soil and gravel without installing a cargo compartment cover and other vehicles that are likely to fall due to poor loading and may impose a fine for negligence.

Prevents traffic accidents that can occur when various loads loaded in the loading box of poorly loaded vehicles, especially large trucks, fall on the road, and the manpower invested to clean up the debris that has fallen to the road surface and the social costs resulting from this The purpose is to reduce

Unlike general roads that run at low speeds, in the case of highways, most vehicles travel at high speeds, so that loads falling from a poorly loaded vehicle that advances ahead of their own vehicle collide with the windshield of the vehicle and are damaged, or debris falling on the road surface Accidents in which vehicle tires are damaged by this are frequently occurring.

In order to reduce the damage to human life and property caused by such a traffic accident and the social cost accompanying it, we are cracking down on vehicles with poor loading in accordance with the aforementioned Article 39 of the Road Traffic Act. Nevertheless, the conventional crackdown on vehicles with poor loading is conducted by simple visual observation of the officers, so the standards for judging vehicles with bad loading are different for each regulator, and this leads to the social action on crackdown on vehicles with bad loading. There is a problem in that reliability is lowered.

It is possible to provide a method and system for detecting illegally loaded vehicles by combining a camera image (front image information) and sensor information.

By inputting the camera image information into the deep learning network, the vehicle model and the vehicle model are distinguished, and the vehicle shape is rendered using the vehicle model and the loading box image information corresponding to the vehicle model extracted based on the vehicle information stored in the database and the lidar sensor. It is possible to provide a method and system for detecting illegally loaded vehicles using information.

It is possible to provide a method and system for detecting an illegally loaded vehicle from the vehicle loading box image information and the vehicle rendering information input to the learned learning model in order to crack down on the poorly loaded truck.

The enforcement system for cracking down on a truck with bad loading includes: a vehicle information acquisition unit configured to acquire rendering information of a vehicle generated using image information of a loading box of a vehicle photographed using a camera and sensor information; an input unit for inputting the acquired vehicle image information and the acquired vehicle rendering information into a learned learning model to crack down on a poorly loaded truck; and an illegal loading detection unit for detecting an illegally loaded vehicle from the vehicle's loading box image information and vehicle rendering information for cracking down on the poorly loaded truck input to the learned learning model.

The vehicle information obtaining unit, a camera and a sensor are mounted on the door-shaped structure of the road, using the camera mounted on the door-shaped structure of the road to obtain the front image information of the vehicle, based on the obtained front image information of the vehicle and generating first profile information, wherein the first profile information includes at least one or more of a vehicle number, a vehicle logo, a vehicle headlight position and shape, or a radiator grill based on the obtained front image information of the vehicle. may include

The vehicle information acquisition unit inputs the first profile information generated based on the front image information of the vehicle into a learned learning model to determine the vehicle model and vehicle model, and selects the vehicle model and model using the learned learning model. can judge

The vehicle information acquisition unit distinguishes a vehicle type including a passenger car, a bus, and a truck from the acquired front image information, recognizes a manufacturer through a logo included in the first profile information, and includes The vehicle model may be distinguished through the position and shape of the headlamp and the radiator grill, and the type of truck may be distinguished through recognition of the vehicle number included in the first profile information.

The vehicle information obtaining unit may extract vehicle identification information corresponding to the determined vehicle type and model using vehicle information stored in a database, and obtain loading box image information including the extracted vehicle identification information.

The vehicle information acquisition unit generates second profile information using a lidar (LiDAR) among sensors mounted on the door-shaped structure of the road, and adjusts the scale of the generated second profile information to provide vehicle rendering information related to the vehicle shape. and generating, wherein the second profile information may include vehicle shape information including vehicle speed, height, and length.

A crackdown method for cracking down on a badly loaded truck includes: acquiring rendering information of a vehicle generated using image information of a loading box of a vehicle photographed using a camera and sensor information; using the obtained image information of the vehicle and the obtained rendering information of the vehicle to input into a learned learning model to crack down on a poorly loaded truck; And it may include the step of detecting the illegally loaded vehicle from the vehicle loading box image information and the vehicle rendering information for cracking down on the badly loaded truck input to the learned learning model.

In the obtaining step, a camera and a sensor are mounted on the door-shaped structure of the road, and the front image information of the vehicle is obtained using the camera mounted on the door-shaped structure of the road, and based on the obtained front image information of the vehicle generating first profile information, wherein the first profile information is based on the acquired front image information of the vehicle, at least one of a vehicle number, a logo of the vehicle, a position and shape of a headlight of the vehicle, or a radiator grill may include

The acquiring may include inputting the first profile information generated based on the front image information of the vehicle into a learning model learned to determine the vehicle model and vehicle model, and selecting the vehicle model and model using the learned learning model. It may include a step of judging.

In the acquiring step, a vehicle model including a passenger car, a bus, and a truck is distinguished from the acquired front image information, a manufacturer is recognized through a logo included in the first profile information, and the It may include the step of classifying the vehicle model through the position and shape of the headlamp and the radiator grill, and classifying the truck type through the recognition of the vehicle number included in the first profile information.

The obtaining may include extracting vehicle identification information corresponding to the determined vehicle type and model using vehicle information stored in a database, and obtaining loading box image information including the extracted vehicle identification information. have.

The acquiring includes generating second profile information using a lidar (LiDAR) among sensors mounted on the door-shaped structure of the road, and adjusting the scale of the generated second profile information to provide vehicle rendering information related to the vehicle shape. and generating, wherein the second profile information may include vehicle shape information including vehicle speed, height, and length.

By combining camera image information and sensor information, the vehicle model and model name can be identified, and the vehicle information stored in the database can be used to more accurately detect a vehicle with poor loading.

The detection performance may be improved by detecting a poorly loaded vehicle from the vehicle's loading box image information and the vehicle's rendering information using the learned model to crack down on the poorly loaded truck.

1 is a view for explaining the general operation of the enforcement system according to an embodiment.
2 is a block diagram illustrating a configuration of an enforcement system according to an embodiment.
3 is a flowchart for explaining a method of cracking down on a badly loaded truck in the enforcement system according to an embodiment.
4 is a view for explaining obtaining front image information of a vehicle using a camera in the enforcement system according to an embodiment.
5 is a diagram for explaining obtaining rendering information of a vehicle by using a lidar sensor in the enforcement system according to an embodiment.
6 is a view for explaining the detection of an illegally loaded vehicle by combining front image information of the vehicle and rendering information of the vehicle in the enforcement system according to an embodiment.
7 is a view for explaining a detailed operation of the enforcement system according to an embodiment.
8 is a view for explaining the detection of a loading faulty truck in the enforcement system according to an embodiment.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings.

1 is a view for explaining the general operation of the enforcement system according to an embodiment.

The enforcement system may detect a vehicle with poor loading (eg, a truck) by combining image information and sensor information using a camera and a sensor mounted on a structure on the road. The enforcement system may be operated on structures installed on the road. For example, the structure may mean a gate-type structure installed on a road to control the speed of a vehicle.

At least one sensor 110 and a camera 120 may be disposed at a central point of an inverted U-shaped frame structure installed on both sides of a multi-road. In this case, various sensors such as a laser sensor, a radar sensor, an IR sensor, etc. may be used, but in the embodiment, a lidar sensor (LiDAR: Light Detection and Ranging) will be described as an example. The lidar sensor irradiates a laser beam to the ground or the atmosphere, collects and analyzes reflected or scattered light from the ground with a telescope, and thereby obtains vehicle-related information, such as the length of the vehicle, the height of the vehicle, the speed of the vehicle, etc. It can indicate the sensor used. For example, the lidar sensor may be installed on both sides of a multi-lane consisting of several lanes to sense the vehicle, and collects a laser beam reflected from the ground or vehicle by irradiating a laser beam from the sky above the vehicle to the ground. , vehicle sensing for generating information such as height and length of the vehicle based on the collected laser beam may be performed. In an embodiment, the lidar sensor may be mounted on a door-shaped structure on the road together with the camera, and in addition to the lidar sensor, a radar, laser, IR sensor, etc. may be additionally attached.

For example, the lidar sensor 110 may be disposed at the upper center point of the inverted U-shaped frame structure installed on both sides of the upward four-lane road. As such, as the lidar sensor 110 is disposed on the top of the structure, the lidar sensor 110 may irradiate a laser beam to the ground and vehicle for each lane from the sky of a vehicle that stops at each lane or passes through each lane. Then, the irradiated beam may be reflected or scattered from the ground or vehicle, and the lidar sensor 110 may collect the reflected or scattered beam. In addition, the camera 120 may photograph a vehicle in real time, or may photograph a vehicle every preset time, or when a vehicle is detected in a camera-capable range at a location where the camera is installed in a structure, the detected vehicle can be filmed

The enforcement system may use the camera 120 image and the lidar sensor 110 to crack down on a vehicle with poor loading. Referring to FIG. 7 , it is a view for explaining a detailed operation of the enforcement system. The enforcement system may acquire front image information using a camera ( 701 ). The enforcement system may generate ( 702 ) the first profile information based on the front image information obtained using the camera. In this case, the first profile information may include at least one of a vehicle number, a vehicle logo, a position and shape of a headlamp of the vehicle, or a radiator grill based on front image information of the vehicle. In other words, the first profile information may be to acquire characteristic information of the vehicle. For example, the vehicle number on the front of the vehicle is different depending on the general car or special car (lorry/truck).

The enforcement system may input the generated first profile information to the learned deep learning-based learning model 703 to determine the vehicle model and model. For example, the enforcement system may build a deep learning-based learning model to determine the vehicle type and model. In this case, the deep learning-based learning model may be composed of a time-based network, for example, LSTM, RNN, and the like. The enforcement system may train a learning model constructed in order to determine the vehicle type and model by using a pre-prepared learning data set. For example, the pre-prepared training data set may be image information for determining a vehicle model and model. The enforcement system may determine ( 704 ) the car model and model from the first profile information using the learned learning model. For example, the enforcement system may distinguish a passenger car, a truck, a bus, etc. using front image information, and may recognize a manufacturer and a model as it is determined to be a truck. The enforcement system recognizes the manufacturer through the logo on the front of the vehicle, distinguishes the vehicle model through the position and shape of the front headlight, the radiator grill, etc., and can distinguish the type of truck through recognition of the vehicle number. The enforcement system may extract vehicle identification information corresponding to the determined vehicle type and model by using the vehicle information 705 stored in the database. In detail, vehicle information on the vehicle, for example, the capacity of the vehicle for each vehicle type and model, the size of the vehicle, the shape of the vehicle, the type of vehicle replacement, etc. may be stored in the database. The enforcement system may extract 2015 model A manufacturer, 5-ton cargo truck, etc. based on the determined vehicle type and model. Furthermore, the enforcement system may acquire characteristic information about the vehicle type and model determined through vehicle information stored in the database. The enforcement system may obtain the loading box image information 706 including vehicle identification information. In this case, the loading box image information may mean front image information including vehicle identification information extracted with respect to the vehicle model and model determined using vehicle information stored in the database. Referring to FIG. 4 , it is a diagram for explaining obtaining image information of a loading box using the camera 120 . Various types of trucks such as cargo, trailer, and trailer may exist, and among them, trucks may be classified in detail according to capacity and size.

The enforcement system may generate ( 711 ) the second profile information using the lidar sensor. Referring to FIG. 5 , it is a diagram for explaining obtaining vehicle rendering information using the lidar sensor 110 . The enforcement system may acquire the second profile information through the lidar sensor 110 . In the enforcement system, the lidar sensor 110 may acquire second profile information including the length of the vehicle, the height of the vehicle, the speed, and the like. For example, the lidar sensor 110 may sense the start time and end time of the vehicle. The enforcement system may acquire vehicle-related information including a start time and an end time sensed from the lidar sensor 110 . Then, the enforcement system may calculate the speed of the vehicle using the start time and the end time.

The enforcement system may perform a matching operation between the second profile information related to the vehicle and a predefined reference profile to generate vehicle shape rendering information. In this case, if the scale (scale) between the vehicle-related second profile and the reference profile is different, comparison may not be possible, so the scale of the second profile may be adjusted ( 712 ) based on the reference profile. Here, the reference profile may represent data obtained by pre-measuring the length, height, speed, etc. of the corresponding vehicle for each vehicle type more than a predefined number of times and then leveling the data. As an example, the enforcement system may perform a comparison and contrast operation between the second profile generated based on the information received from the lidar sensor 110 and a predefined reference profile. In this case, a scale between two profiles, for example, a resolution between images corresponding to each of the two profiles, may be different according to a sampling rate, a moving speed of a vehicle, etc. during the comparison and contrast operation. Accordingly, if the reference profile is measured and compared for each unit speed of the vehicle, the resource is wasted severely, so after the vehicle passes the detection area, the vehicle length is divided into equal intervals regardless of the vehicle speed, The scale of the second profile information can be adjusted by setting the number of data in the .

Depending on the vehicle model, the height of the load and the shape of the load box may be different. The enforcement system may render 713 the shape of the vehicle based on the height of the load. In this case, the enforcement system may divide the vehicle into a total of N sections in order to compensate for the rendering information of the vehicle that has rendered the shape of the vehicle, and calculate the partial area of the vehicle by using the vehicle length and the height of each section of the vehicle.

Referring to FIG. 6 , it is an example for explaining detection of an illegally loaded vehicle by combining front image information of the vehicle and rendering information of the vehicle. The enforcement system may use the loading box image information 660 and the rendering information 630 to input into the learned learning model to control the badly loaded truck. At this time, the loading box image information 660 is generated using the front image information 610 of the vehicle, and the rendering information 630 may mean the shape rendering information of the vehicle in which the shape of the vehicle is rendered using the lidar sensor. have. For example, enforcement systems can build deep learning-based learning models to crack down on poorly loaded trucks. In this case, the enforcement system may build a learning model for cracking down on a badly loaded truck separately from the learning model for determining the vehicle type and model described above. Similarly, a deep learning-based learning model for cracking down on badly loaded trucks may be composed of a time-based network, for example, LSTM, RNN, and the like. The enforcement system can train a deep learning-based learning model by using the training data for cracking down on poorly loaded trucks. For example, the training data may be image information including truck information. In addition, a deep learning-based learning model for cracking down on poorly loaded trucks in various ways can be learned.

The enforcement system may detect an illegally loaded vehicle from the loading box image information 660 of the vehicle and the rendering information 630 of the vehicle for cracking down on the badly loaded truck using the learned learning model. The enforcement system can determine whether the vehicle is loaded badly from the image information 660 of the vehicle and the rendering information 630 of the vehicle using the learned learning model for the loading box image information 660 and the vehicle rendering information 630. have. For example, information on the loading box of the vehicle (eg, loading box type, size, capacity, etc.) according to the vehicle type and model of the vehicle may be set. The enforcement system can recognize a vehicle with poor loading if it has a different shape from the loading box information. The enforcement system may determine the recognized badly loaded vehicle as an illegally loaded vehicle. In addition, the enforcement system may estimate the degree of loading failure of the vehicle determined to be a vehicle with poor loading. Referring to FIG. 8 , it is an example showing a truck detected as a loading defect. For example, when the enforcement system detects that an object larger than the size of the truck's loading box is loaded and moved, it may be determined as a poorly loaded truck. In other words, when an item larger than the size or capacity of the load box is loaded or the load box is loaded in excess of the capacity, as another example, the enforcement system is to move without locking the item in the load box of the truck. If it detects that it is moving without a mesh, it can be judged as a poorly loaded truck.

FIG. 2 is a block diagram illustrating the configuration of an enforcement system according to an embodiment, and FIG. 3 is a flowchart illustrating a method of cracking down on a badly loaded truck in the enforcement system according to an embodiment.

The enforcement system 100 may include a vehicle information acquisition unit 210 , an input unit 220 , and an illegal loading detection unit 230 . The components of the enforcement system 100 may be representations of different functions performed by the processor according to instructions provided by the program code stored in the enforcement system 100 . The enforcement system 100 and components of the enforcement system 100 may control the enforcement system to perform steps 310 to 330 included in the method of cracking down on a poorly loaded truck of FIG. 2 . In this case, the enforcement system 100 and the components of the enforcement system 100 may be implemented to execute instructions according to the code of the operating system included in the memory and the code of at least one program.

The processor may load into memory the program code stored in the file of the program for a method of cracking down on a badly loaded truck. For example, when the program is executed in the enforcement system, the processor may control the enforcement system to load the program code from a file of the program into the memory according to the control of the operating system. At this time, the enforcement system 100 and the vehicle information acquisition unit 210 , the input unit 220 , and the illegal loading detection unit 230 included in the enforcement system 100 each receive a command of a corresponding part of the program code loaded in the memory. may be different functional representations of the enforcement system 100 for executing and subsequently executing steps 310 - 330 .

In step 310 , the vehicle information obtaining unit 210 may obtain rendering information of the vehicle generated using the image information of the loading box of the vehicle photographed using the camera and the sensor information. The vehicle information acquisition unit 210 is equipped with a camera and a sensor on the door-shaped structure of the road, acquires front image information of the vehicle using the camera mounted on the door-shaped structure of the road, and based on the obtained front image information of the vehicle First profile information may be generated. The vehicle information acquisition unit 210 inputs the first profile information generated based on the front image information of the vehicle to the learned learning model to determine the vehicle model and vehicle model, and uses the learned learning model to select the vehicle model and model. can judge Vehicle information acquisition unit 210 distinguishes a vehicle including a passenger car, a bus, and a truck from the acquired front image information, recognizes the manufacturer through the logo included in the first profile information, and controls the headlamp included in the first profile information. The vehicle model can be distinguished through the location and shape and the radiator grill, and the type of truck can be distinguished through the recognition of the vehicle number included in the first profile information. The vehicle information obtaining unit 210 may extract vehicle identification information corresponding to the determined vehicle type and model by using vehicle information stored in the database, and obtain loading box image information including the extracted vehicle identification information. The vehicle information acquisition unit 210 generates second profile information by using a lidar (LiDAR) among sensors mounted on door-shaped structures on the road, and renders a vehicle related to a vehicle shape through scale adjustment of the generated second profile information. information can be generated.

In step 320 , the input unit 220 may input the obtained image information of the vehicle and the obtained rendering information of the vehicle to the learned learning model in order to crack down on a poorly loaded truck.

In step 330 , the illegally loaded detection unit 230 may detect an illegally loaded vehicle from the image information of the loading box of the vehicle for cracking down on the badly loaded truck input to the learned learning model and the rendering information of the vehicle.

The device described above may be implemented as a hardware component, a software component, and/or a combination of the hardware component and the software component. For example, devices and components described in the embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA). , a programmable logic unit (PLU), microprocessor, or any other device capable of executing and responding to instructions, may be implemented using one or more general purpose or special purpose computers. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to execution of the software. For convenience of understanding, although one processing device is sometimes described as being used, one of ordinary skill in the art will recognize that the processing device includes a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that may include For example, the processing device may include a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as parallel processors.

The software may comprise a computer program, code, instructions, or a combination of one or more thereof, which configures a processing device to operate as desired or is independently or collectively processed You can command the device. The software and/or data may be any kind of machine, component, physical device, virtual equipment, computer storage medium or device, to be interpreted by or to provide instructions or data to the processing device. may be embodied in The software may be distributed over networked computer systems and stored or executed in a distributed manner. Software and data may be stored in one or more computer-readable recording media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the embodiment, or may be known and available to those skilled in the art of computer software. Examples of the computer-readable recording medium include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic such as floppy disks. - includes magneto-optical media, and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like.

As described above, although the embodiments have been described with reference to the limited embodiments and drawings, various modifications and variations are possible by those skilled in the art from the above description. For example, the described techniques are performed in a different order than the described method, and/or the described components of the system, structure, apparatus, circuit, etc. are combined or combined in a different form than the described method, or other components Or substituted or substituted by equivalents may achieve an appropriate result.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

Claims (12)

In the enforcement system for cracking down on defective trucks,
a vehicle information acquisition unit configured to acquire rendering information of a vehicle generated by using the image information of the loading box of the vehicle photographed using the camera and the sensor information;
an input unit for inputting the acquired vehicle image information and the acquired vehicle rendering information into a learned learning model to crack down on a poorly loaded truck; and
An illegal loading detection unit that detects an illegally loaded vehicle from the vehicle's loading box image information and vehicle rendering information for cracking down on the poorly loaded truck input to the learned learning model
including,
The vehicle information acquisition unit,
Obtaining front image information of a vehicle using a camera mounted on a door-shaped structure of a road, generating first profile information based on the obtained front image information of the vehicle, and generating the generated front image information of the vehicle The first profile information is input into the learned learning model to determine the vehicle model and vehicle model, the vehicle model and model are determined using the learned learning model, and the determined vehicle model and model are determined using the vehicle information stored in the database. Extracting vehicle identification information corresponding to , obtaining loading box image information including the extracted vehicle identification information, and generating second profile information using a lidar (LiDAR) among sensors mounted on door-shaped structures on the road, Generate vehicle rendering information related to a vehicle shape through scale adjustment of the generated second profile information,
The scaling includes adjusting the scale of the second profile based on the reference profile through a matching operation between the second profile information and a predefined reference profile,
The reference profile includes data obtained by measuring the length, height, and speed of the corresponding vehicle by a predetermined number of times or more in advance for each vehicle type and leveling it.
Enforcement system characterized by. According to claim 1,
Including a camera and a sensor mounted on the door-shaped structure of the road,
The first profile information includes at least one of a vehicle number, a vehicle logo, a vehicle headlight position and shape, or a radiator grill based on the acquired front image information of the vehicle
Enforcement system, characterized in that. delete According to claim 1,
The vehicle information acquisition unit,
Differentiate a vehicle including a passenger car, a bus, and a truck from the obtained front image information, recognize a manufacturer through a logo included in the first profile information, and position and shape of a headlamp included in the first profile information, Classifying the vehicle model through the radiator grill and classifying the truck type through the recognition of the vehicle number included in the first profile information
Enforcement system, characterized in that. delete According to claim 1,
The second profile information includes shape information of the vehicle including the speed, height, and length of the vehicle.
Enforcement system, characterized in that. In the crackdown method for cracking down on a badly loaded truck,
obtaining rendering information of the vehicle generated using the image information of the loading box of the vehicle photographed using the camera and the sensor information;
using the obtained image information of the vehicle and the obtained rendering information of the vehicle to input into a learned learning model to crack down on a poorly loaded truck; and
Detecting an illegally loaded vehicle from the vehicle's loading box image information and vehicle rendering information for cracking down on the badly loaded truck input to the learned learning model
including,
The obtaining step is
Obtaining front image information of a vehicle using a camera mounted on a door-shaped structure of a road, generating first profile information based on the obtained front image information of the vehicle, and generating the generated front image information of the vehicle The first profile information is input into the learned learning model to determine the vehicle model and vehicle model, the vehicle model and model are determined using the learned learning model, and the determined vehicle model and model are determined using the vehicle information stored in the database. Extracting vehicle identification information corresponding to , obtaining loading box image information including the extracted vehicle identification information, and generating second profile information using a lidar (LiDAR) among sensors mounted on door-shaped structures on the road, Generating vehicle rendering information related to a vehicle shape through scale adjustment of the generated second profile information
including,
The scaling includes adjusting the scale of the second profile based on the reference profile through a matching operation between the second profile information and a predefined reference profile,
The reference profile includes data obtained by measuring the length, height, and speed of the corresponding vehicle by a predetermined number of times or more in advance for each vehicle type and leveling it.
Control method characterized in that. 8. The method of claim 7,
Including a camera and a sensor mounted on the door-shaped structure of the road,
The first profile information includes at least one of a vehicle number, a vehicle logo, a vehicle headlight position and shape, or a radiator grill based on the acquired front image information of the vehicle
Control method, characterized in that. delete 8. The method of claim 7,
The obtaining step is
Differentiate a vehicle including a passenger car, a bus, and a truck from the obtained front image information, recognize a manufacturer through a logo included in the first profile information, and position and shape of a headlamp included in the first profile information, Classifying a vehicle model through a radiator grill, and classifying a truck type through vehicle number recognition included in the first profile information
Enforcement methods, including. delete 8. The method of claim 7,
The second profile information includes shape information of the vehicle including the speed, height, and length of the vehicle.
Control method, characterized in that.

Images