KR20090030666A - High-speed weight in motion - Google Patents

Abstract

An automatic weight measurement system for high speed vehicles is provided to detect the overload of high speed vehicles by using an automatic license plate recognition device, a lane change and vehicle trace unit and a 3D length measuring unit and a vehicle width measuring unit. An automatic weight measurement system for high speed vehicles comprises: an automatic license plate recognition device(131) recognizing a license plate numbers from the vehicle image information photographed through a first camera; a lane change and vehicle trace unit(138) tracing location and movement route of the vehicle from the vehicle image photographed from a second camera; a 3D length measuring unit(134) and a vehicle width measuring unit(135) determining overload of a cargo and vehicle type sorting based on the vehicle image obtained through the first and the second cameras; a speed sensor(137) detecting vehicle velocity; and a vehicle sorting and overload detecting unit(136) classifying the vehicle model of the vehicle and detecting the overload by using license plate information, 3D length information and vehicle width information.

Description

High-speed Weight In Motion System

The present invention is to enable the automatic measurement of the grade and appearance size of the vehicle running at high speed by the non-contact measurement method through the 3D imaging device.

Due to the continuous development of national industrialization and the increase in the volume of transportation, the budget for the maintenance of the base road network has been increased. In particular, vehicle overloading is a major factor that causes damage to the base road, and has been proposed a vehicle relaying system that measures the weight of the vehicle and if the vehicle is overestimated, it adds a fine.

The vehicle weighting system proposed in the related art measures the weight of a vehicle when the vehicle is stopped or measures the weight of the vehicle below a specific speed.

Therefore, such a conventional vehicle weighing system induces traffic congestion due to vehicle induction and stopping because the vehicle must be measured at a stationary state or below a certain speed, and in particular, each axis weight must be summed to measure the total load. There was a disadvantage that the measurement error is severe according to the measurement method.

In addition, there is a disadvantage in that a specific area for vehicle traffic must be secured on the back of the road, causing traffic accidents in the process of inducing the vehicle to the location of the equipment, and there is a disadvantage in that the vehicle ignores the request for vehicle guidance.

In addition, the conventional vehicle weighting system that embeds a piezo sensor or a loop sensor on the road surface and induces the measurement system to pass through the measurement system at a certain speed or less is not required to incur unnecessary costs due to the decrease in traffic flow caused by the speed decrease. The maintenance cost exceeded the maintenance cost, and there was no ability to cope with rapid lane change and rapid deceleration, especially to avoid measurement, and also provided a cause of an accident.

As a countermeasure against this, developed countries are conducting research on developing high speed weight in motion (HWIM) suitable for each country's environment.

The present invention has been proposed in order to solve all the problems that occur in the conventional vehicle relay system as described above,

An object of the present invention is to provide a high-speed driving vehicle automatic weighing system capable of automatically measuring the real-time rating and the appearance size of the high-speed driving vehicle by a non-contact measurement method through the 3D imaging device.

Another object of the present invention is to provide a high-speed driving vehicle automatic weighing system capable of coping with a measurement avoidance operation when entering a measurement system of a measurement target vehicle and enabling non-contact measurement of the volume (length and vehicle width) of the vehicle.

It is still another object of the present invention to provide a high-speed driving vehicle automatic weighing system capable of tracking the flow of a vehicle through a non-contact image detector and correcting the information of the measured sensor based on the position of the sensor and the moving path of the vehicle. have.

It is another object of the present invention to accumulate all information on the weight, speed, volume, etc. of all vehicles passing through the road, and to re-process these information to drive efficient high-speed vehicles that can promote efficient maintenance strategies. It is to provide a relay system.

"High speed vehicle automatic weighing system" according to the present invention,

When the entry of the vehicle is detected by the vehicle entry / exit detection unit embedded in the road for the vehicle to detect the entry and exit of the specific line, the weight sensing device measures the total gross weight of the entered vehicle.

The vehicle recognition and reading device is composed of a 3D CCD camera, a character recognition and reading module, and extracts the speed information of the vehicle entering the specific line, extracts license plate information from the image information, and extracts the vehicle number by analyzing it. In addition, using a stereo vision system to analyze the vehicle grade and appearance size from the image information of the vehicle in a non-contact manner.

The controller effectively analyzes and combines the entrance information, the weight information, and the vehicle appearance measurement information acquired from the vehicle entry / exit detection unit, the weight detection device, and the vehicle recognition / reading device, respectively, to determine the vehicle classification and the volume overload. In the event of overload, a warning message is displayed on the display.

According to the present invention, it is possible to detect the overloaded state of the vehicle, it can be utilized to prevent the entry of a specific area of the overloaded vehicle or accusations of violation of the law, and to use the video / steel cameras installed in the system as an active overload control device. As an accusation or passive approach, the HWIM station is installed several kilometers before the existing overload vehicle control point, allowing the detected vehicle to be screened without interruption to traffic flow and called as a precision measurement point for inspection of the vehicle. And static weighing or by-pass road guidance.

In addition, by preventing the entry of the bridge overload vehicle overload weight, there is an advantage that can be prevented in the future bridge breakage and maintenance obstacles in advance, the number of vehicles passed per hour, the vehicle type, the presence of cargo, the number of cargo vehicles, It can collect accurate statistical data about freight class, vehicle speed, passing vehicle interval, etc., and can help to analyze traffic accurately based on this. Also, it is possible to implement differential tariff based on vehicle weight. As it is possible to accurately classify according to the loading weight as well as the overload, it has the advantage of allowing detailed charges according to the loading weight, beyond the general toll charging method according to the three levels.

Furthermore, the present invention has the advantage of implementing the optimized road management, diagnosis, and maintenance by setting up a decision-making process applied with HWIM, and constructing various accumulated data in an optimized database, and building road conditions and structures. By providing historical information according to traffic volume and traffic volume, it is possible to predict future traffic changes and maintenance requirements, and to calculate effective road maintenance time and processing cost. Based on this, important basic information about new road construction It also has the advantage of helping to derive

The traveling vehicle automatic monitoring system according to the present invention has an advantage of automatically measuring the real-time rating and the outline size of the high-speed driving vehicle by a non-contact measurement method.

In addition, there is an advantage that it is possible to cope with the measurement avoidance start when entering the measurement system of the vehicle to be measured and non-contact measurement of the volume (length and vehicle width) of the vehicle.

In addition, the non-contact image detector has the advantage of tracking the flow of the vehicle and correct the information of the weight sensor measured based on the position of the weight sensor and the movement path of the vehicle.

In addition, it is possible to accumulate all information on the weight, speed, volume, etc. of all vehicles passing through the road, and to reprocess the information, thereby promoting an efficient maintenance strategy.

Hereinafter, described in detail with reference to the accompanying drawings a preferred embodiment of the present invention. If it is determined that the detailed description of the known function or configuration related to the present invention may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted.

1 is a schematic diagram schematically showing a system for automatically driving a high speed vehicle according to an exemplary embodiment of the present invention, and FIG. 2 is a block diagram illustrating a system for automatically starting a high speed vehicle according to FIG. 1. 110, a weight sensing device 120, a vehicle recognition and reading device 130, a display device 140, and a controller 150.

The vehicle entry / exit detection unit 110 detects the vehicle entry and exit of a specific line of the road through a change in inductance of the loop coil using the loop coil sensors 111 and 112 to detect the vehicle to be measured. Be aware of entry and exit.

The weight sensing device 120 detects an axial load for each wheel of the driving vehicle by using a piezo crystal sensor which is a load sensing sensor 121 and extracts a total weight signal of the entire measurement vehicle.

The vehicle recognition and reading device 130 is composed of a 3D CCD camera 131 and a character recognition and reading software (S / W) module to extract the speed information of the vehicle entering the specific line and to obtain the license plate from the image information. The information is extracted, the vehicle number is extracted by analyzing the information, and the vehicle grade and the size of the vehicle are analyzed in a non-contact manner using the stereo vision system.

The vehicle recognition and reading device 130 acquires a three-dimensional image of the vehicle by applying stereo vision recognition technology using two CCD cameras, recognizes the size of the exterior, the lane entry angle, the length, and the like to determine whether the sensor passes through the predetermined lane. And volume overload, i.e. exceeding lane width or violating a defined total allowable length.

The display device 140 notifies the overloaded vehicle driver of warnings and further measures related to the overload violation so as to guide the overloaded vehicle to a specific area for departure from the corresponding road or detailed overweight measurement.

The controller uses the gross weight of the measured vehicle, vehicle identification and readout information to determine the application of road traffic regulations. The controller additionally includes a communication module that can send data to the central control computer, providing a search function to accumulate data in the central computer database and retrieve vehicle information.

For reference, the above-described central control system stores and manages vehicle weight information obtained from each HWIM (automatic vehicle speed monitoring system) through a systematic storage management system (DBMS), and can generate knowledge therefrom. Provide knowledge management system for road network management by establishing knowledge information system. It also manages and controls HWIM system operation information.

The present invention detects the entry of the measurement target vehicle by the loop coil sensor 111 of the vehicle entry / exit detection unit 110 when the measurement target vehicle travels on the road on which the high-speed driving vehicle automatic relay system is installed. . Here, the loop coil sensor 111 detects a vehicle entry of a road specific line using a change in inductance of the loop coil.

When the vehicle to be measured moves forward after the vehicle entry is detected, the weight of the vehicle to be measured is started by the load sensor 121 of the weight sensing device 120. Here, the load sensor 121 is buried in an appropriate number effective for a specific vehicle line, as shown in FIG.

Piezoelectric sensors, piezoelectric film sensors and the like are utilized as basic sensors for measuring the axial weight of a vehicle. In the present invention, piezoelectric crystals are used as basic sensors of the system. The weight sensing device 120 using such a sensor converts the mechanical load applied to the crystal piece into electrical signals by releasing charges according to the load direction when mechanical pressure is applied to measure the magnitude thereof. Provide a measurement tool.

When the wheels of the vehicle pass through the pad of the load sensor 121, it is preferable to design the sensor so that the surface of the sensor pad does not deform even when vertical force is applied, and discharges electrons by the mechanical pressure applied by the tire. Convert electrical signals proportional to the load.

The electrical signal obtained from the load sensor 121 is efficiently converted into a voltage signal using a charge amplifier, integrated through an integrator, and then amplified to a predetermined level using an inverting amplifier. Thereafter, the amplified analog detection signal is converted into a digital signal to calculate the weight of one axis, and the weight calculated for each axis is added to the controller 150 as the total weight of the vehicle.

Next, when the vehicle enters the measurement target area, the vehicle recognition and reading apparatus 130 captures the vehicle image with the two CCD cameras 131, and processes and analyzes the captured image information to determine whether it is overloaded. The necessary information is extracted and provided to the controller 150.

The information measured by the HWIM system is processed based on the information input at the position of the road lane in which the load sensor 121 of the weight sensing device 120 is embedded. However, when the vehicle to be measured changes lanes rapidly at the location of the load sensor 121 installed, the existing system has an error that causes confusion between the information measured in each lane. It was impossible. In addition, the influence of speed is very important among the important parameters for analyzing the input load measurement signal. However, in order to avoid this, there is no suggestion to effectively cope with the rapid speed deceleration and increase when passing through the measurement sensor.

On the other hand, in the present invention, in order to solve the problems that occur in the conventional vehicle monitoring system as described above, the flow of the vehicle is tracked through a non-contact image detector and the load detection based on the position of the load sensor and the movement path of the vehicle The sensor's information can be compensated. In addition, it collects various information of the vehicle and can be used as a supplement to the weight information collected through the HWIM sensor.

In more detail, it is determined whether the overload by the data of the measured weight sensing device to determine the enforcement target. In order to control the overestimated vehicle when it is determined by the weight sensing device, the license plate of the vehicle is recognized and the driver of the overloaded vehicle is notified of the follow-up action or the control information is displayed through the front light guide plate.

As shown in FIG. 3, the license plate of the vehicle is photographed through a separate CCD camera 131 to recognize the license plate of the vehicle. The photographed image is transferred to an automatic visual identification (AVI) device 132, and the license plate area extraction unit 132a extracts the license plate area as shown in FIG. 4 based on the luminance change in the character area. The license plate information is recognized through binarization of the extracted region.

The vehicle number position separating unit 132b separates the location of the area code, the use classification, and the vehicle number from the recognized license plate information, and the character recognition unit 132c performs character recognition on each of the divided regions as shown in FIG. 5C. The license plate information is converted into Korean code and numbers through the license plate information and transmitted to the vehicle type classification and overload detection unit 136.

The vehicle type classification and overload detection unit 136 transmits the corresponding information to the controller 150 and registers it in the vehicle information database 151. The overload determination unit 153 of the controller 150 transfers the information obtained through the AVI 132 to the central control database so that the record of the vehicle can be retrieved through a DB search. In addition, it is possible to collect important road information such as hourly traffic volume and passing vehicle grade, speed, and total cumulative load through image signal analysis.

Next, in order to avoid crackdown in the overload vehicle control system using the HWIM sensor installed on the road surface, the lane may be changed by changing the lane at the location where the weight sensor is installed. Therefore, in the present invention, the vehicle lane change and vehicle tracking unit 133 tracks the position and the movement path of the vehicle from the image input through the CCD camera 138 to calculate the position of the weight detection sensor passed by the vehicle, and from there The weight information detected by the weight sensor is corrected with information of the vehicle in which the lane change has occurred.

To this end, the position of the weight sensor installed on the road surface specifies the position and direction on the screen of the input camera and sets the position of each weight sensor. The lane area extractor 133a extracts the vehicle area in which the movement occurs by calculating the input screen and the background screen excluding the vehicle every frame in the continuous input screen.

The vehicle movement path tracking unit 133b detects the vehicle region extracted every frame, tracks the movement route of the vehicle on the previous screen and the current screen, and compares the position of the weight sensor that is set for each region with the unique number of the sensor that passes. Extract The measured value of the weight sensor is searched and corrected to be calculated as the weight of the current vehicle.

6a to 6c show the lane change and the position of the weight sensor of the measurement target vehicle, Figures 7a to 7d is a change graph of the weight data detected by the weight sensor when changing the lane, Figure 7e is the movement of the vehicle It shows the weight sensing data corrected by tracking the path.

Meanwhile, the three-dimensional length measuring unit 134 and the vehicle width measuring unit 135 measure the three-dimensional length and the vehicle width by using the image of the vehicle extracted from the input image and measure the three-dimensional length and the vehicle width to determine whether the overload is loaded. The division and overload detection unit 136 is provided.

That is, the three-dimensional length measuring unit 134 inputs the length and the direction of the reference pattern set on the road surface and the installation position of the camera 138 on the road images input from the two CCD cameras 138 and then in the three-dimensional space. The camera model parameters are calculated using the camera's viewing angle, the angle of the road surface, and the distance, and the actual length is calculated based on the position and length on the screen.

In general, an image appearing on the CCD camera 138 is represented in the form of perspective projection according to the focal length and angle of the lens. When coordinates in three-dimensional space are used in the two-dimensional plane through the conversion of the visual coordinates according to the installation position of the CCD camera 138, the distance between the three-dimensional coordinates and the two-dimensional coordinates through the conversion of the distance between each pixel and the viewing plane A relationship can be obtained.

As shown in FIG. 9, the relationship between the viewpoint of the CCD camera 138 and the projection center line is the same as the process of converting the tilted frustum view volume into a rectangular parallelepiped view volume when performing an orthographic projection. In other words, the relationship between the one point P (x, y, z) in space and the point Q projected on the plane is obtained by calculating the coordinate P (x ', y', z ') when shearing. It can be obtained by calculating the position P (x ", y", z ") corresponding to Q in orthographic projection through scaling).

Shear transformation is performed to find the coordinates when the projection center line is aligned with the Z axis of the viewpoint and projected onto the projection plane of the screen. At this time, the change amount of x and y of the point P is proportional to the distance to the viewpoint O, and the following relationship is established.

here

When the shear transformation is expressed in the form of a matrix from the equation (1) below.

P (x ', y', z ') transformed in the form of orthogonal perspective projection can be converted to orthogonal projection through scaling to obtain a corresponding two-dimensional plane position. The point P (x ', y', z ') in the frustum is closer to the projection center line in proportion to the distance to the viewpoint and z' does not change.

At this time, the original value is maintained as z "= z '= z and Equation 2 below is established.

Therefore, the orthogonal projection transformation is expressed in a matrix as shown in [Equation 3] below.

The relationship between the point P in space at any position and the point Q appearing on the camera's view plane is expressed by Equation 4 below.

의 역행렬을 구함으로써 3차원 공간에서의 좌표값을 구할 수 있다.If you input the start point and the end point of the unit length in the input image, you can get the coordinate values in the 2D plane and in the matrix above

We can find the coordinate values in three-dimensional space by finding the inverse of.

When an arbitrary vehicle image is input, a coordinate value in a three-dimensional space can be obtained from the obtained matrix, and the actual length can be estimated from the distance between the measured pixels.

The vehicle type classification and overload detection unit 136 classifies the vehicle type based on the width and the length of the vehicle detected in the image and calculates the width of the load to determine whether or not to load the excess cargo.

The controller 150 controls the overall system and analyzes information provided from the weight sensing device 120 and the vehicle recognition and reading device 130 to determine whether the vehicle to be measured is overloaded, and the display device 140 when overloading. Print the saved warning message with).

Here, the controller 150 corrects the weight detection data detected by the weight sensing device 120 according to the vehicle movement path tracking value output from the vehicle recognition and reading device 130, and the corrected weight detection data and the vehicle recognition. And finally comparing with the overload information stored in the vehicle information database based on the vehicle number information, the vehicle type classification, and the overload detection information provided by the reading device 130, and warning the display device 140 when overloading. Will print a message.

In this case, the display device 140 may be implemented by a small display means such as an LED, or may be implemented by using a data display device such as an LCD. Such a display device 140 may have a conventional well-known configuration.

The vehicle information database may be any non-volatile memory such as, for example, an EEPROM or a flash ROM, and may be implemented as a high capacity hard disk or a mass storage device capable of network access. The vehicle information database stores vehicle information such as a vehicle type, a vehicle owner, and a vehicle registration place of each vehicle by vehicle number, and vehicle information, vehicle overload information, that is, reference vehicle information based on weight of each vehicle type.

Vehicle information stored in the vehicle information database may be accessed by the vehicle recognition and reading device 130.

Therefore, the automatic driving system of the high speed driving vehicle according to the present invention can determine the overload for each vehicle type of the high speed driving vehicle, and automatically determine whether the vehicle is overloaded or overloaded even if the traffic is congested or the driver ignores the overload vehicle guidance request. Judgment has the advantage that can be processed.

Additionally, the traveling vehicle automatic relay system according to the present invention may store and provide entry / exit history information for each number of vehicles relayed by the system.

Accordingly, the controller 150 may further include a traffic information database for storing information about the entry and exit of the vehicle, and a road traffic information collection unit for storing the entry and exit history information of the vehicle by vehicle number in the traffic information database. .

The passage information database may be any non-volatile memory such as, for example, an EEPROM or a flash ROM, and may be implemented as a high capacity hard disk or a mass storage device having network access. The traffic information database stores information such as speed and weight measured for each traffic time by vehicle number. The road traffic information collecting unit receives information such as speed and weight measured for each traffic time by vehicle number and stores it in the traffic information database, and periodically provides it to a central control center connected through a network so that it can be used for calculating the total traffic volume.

The present invention is not limited to the above-described specific preferred embodiments, and various modifications can be made by any person having ordinary skill in the art without departing from the gist of the present invention claimed in the claims. Of course, such changes will fall within the scope of the claims.

1 is a schematic diagram schematically showing a high speed vehicle automatic monitoring system according to an embodiment of the present invention.

FIG. 2 is a block diagram illustrating a system for automatically counting a high speed vehicle according to FIG. 1.

3 is a block diagram illustrating an example of the vehicle recognition and reading device of FIG. 2;

4 is an exemplary view of a license plate area extracted from a vehicle image.

5A is a binarized screen of the extracted license plate area, FIG. 5B is an area division screen obtained by dividing the extracted license plate area, and FIG. 5C is result data of recognizing the partition area.

6A to 6C are diagrams illustrating lane changes and sensors of a measurement target vehicle.

7A to 7D are weight data detected by the weight sensor according to lane change, and FIG. 7E is weight data corrected by the detected weight data.

8 is a relationship diagram between a camera position and a vehicle image;

9 is a coordinate transformation relationship through perspective projection.

<Explanation of symbols for the main parts of the drawings>

110... Vehicle entry and exit detection unit

120... Weight sensing device

130... Vehicle recognition and reading device

140... Display

150... controller

Claims (8)

An automatic vehicle number recognition device for recognizing a vehicle number from vehicle image information photographed through a first camera; The position of the vehicle and the path of the vehicle are tracked from the image of the vehicle captured by the second camera to calculate the position of the weight sensor passed by the vehicle. A vehicle lane change and vehicle tracking unit configured to calculate vehicle movement path tracking information so as to correct the error; A three-dimensional length measuring unit configured to measure a three-dimensional length for determining a vehicle type classification and whether or not an excessive cargo is loaded by using the image of the vehicle extracted from the vehicle images acquired through the first and second cameras; A vehicle width measuring unit configured to measure a vehicle width for determining a vehicle type classification and whether or not an excess cargo is loaded by using the image of the vehicle extracted from the vehicle images acquired through the first and second cameras; A speed sensor detecting a speed of the vehicle; A vehicle type classification and an overload detection unit which searches for a vehicle information database storing vehicle information by using the recognized vehicle number information, three-dimensional length information, and vehicle width information as search information to classify the vehicle type and detect whether the vehicle is overloaded; High-speed driving vehicle automatic relay system, characterized in that configured. According to claim 1, The automatic vehicle number recognition device, A license plate area extraction unit which extracts a license plate area based on a luminance change in the character area and recognizes license plate information through binarization of the extracted area; A vehicle number location separation unit for separating a location number, a use classification, and a location of a vehicle number from the recognized license plate information; And a character recognition unit for converting each separated area into a Korean code and a number through character recognition to recognize number information. According to claim 1, wherein the vehicle lane change and vehicle tracking unit, A lane area extracting unit for extracting a vehicle area in which a motion is generated by calculating a screen input except a vehicle and a background screen except for a vehicle every frame on a continuously input screen; Detects the vehicle area extracted every frame, tracks the movement path of the vehicle on the previous screen and the current screen, extracts the unique number of the weight sensor passed by comparing with the position of the weight sensor set for each area, and the corresponding weight And a vehicle movement path tracking unit configured to search a measured value of the detection sensor and calculate a weight correction value to correct the weight of the current vehicle. According to claim 3, The vehicle movement path tracking unit, A high-speed driving vehicle automatic weighing system, characterized in that the position and direction of the weight sensor installed on the road surface and the input camera screen, the position of each weight sensor is set in advance, and the weight correction value is calculated. The method of claim 1, wherein the three-dimensional length measuring unit, The length and direction of the reference pattern set on the road surface and the installation position of the camera are input on the road image input from the second camera, and the camera model is constructed using the viewing angle of the camera, the angle of the road surface, and the distance in a three-dimensional space. A high speed driving vehicle automatic weighing system, characterized in that the parameter is calculated and the actual vehicle length is calculated based on the position and length of the screen. A vehicle entry / exit detection unit configured to detect entry and exit of a running vehicle; A weight sensing device configured to detect a total weight according to the speed and load of the vehicle detected by the vehicle entry / exit detection unit; It is equipped with a camera and a text recognition and reading software module, and extracts the speed information of the entered vehicle and acquires a three-dimensional image from the image information of the photographed vehicle to recognize the size, lane entry angle, length of the vehicle, etc. A vehicle recognition and reading device configured to determine whether the sensor portion passes and the volume is overloaded in a predetermined lane; A display device for displaying whether the vehicle is overloaded; The overall system is controlled, the vehicle weight detection data detected by the weight sensing device is corrected according to the vehicle movement path tracking value output from the vehicle recognition reading apparatus, and the overload is determined by determining whether the vehicle is traveling at high speed. A controller configured to output a warning message to the display device; High speed driving vehicle automatic counting system comprising a. The method of claim 6, wherein the controller, And finally determining whether the vehicle is overloaded based on the corrected weight detection data, the vehicle number information provided by the vehicle recognition and reading device, the vehicle type classification, and the overload detection information stored in the vehicle information database. High speed driving vehicle automatic weighing system. The method of claim 6, wherein the vehicle recognition and reading device, An automatic vehicle number recognition device for recognizing a vehicle number from vehicle image information photographed through a first camera; The position of the vehicle and the path of the vehicle are tracked from the image of the vehicle captured by the second camera to calculate the position of the weight sensor passed by the vehicle. A vehicle lane change and vehicle tracking unit configured to calculate vehicle movement path tracking information so as to correct the error; A three-dimensional length measuring unit configured to measure a three-dimensional length for determining a vehicle type classification and whether or not an excessive cargo is loaded by using the image of the vehicle extracted from the vehicle images acquired through the first and second cameras; A vehicle width measuring unit configured to measure a vehicle width for determining a vehicle type classification and whether or not an excess cargo is loaded by using the image of the vehicle extracted from the vehicle images acquired through the first and second cameras; A speed sensor detecting a speed of the vehicle; A vehicle type classification and an overload detection unit which searches for a vehicle information database storing vehicle information by using the recognized vehicle number information, three-dimensional length information, and vehicle width information as search information to classify the vehicle type and detect whether the vehicle is overloaded; High-speed driving vehicle automatic relay system, characterized in that configured.

Images