KR101974105B1 - Photographing system and method for increasing recognition rate of vehicle number - Google Patents

Abstract

The present invention relates to a vehicle photographing system with a higher vehicle number recognition rate and a method thereof. Since a regulated vehicle is photographed at an exact moment when the vehicle passes by a reference spot (L) using the position of the reference spot (L) and the current position and speed of the regulated vehicle, a number plate photographing spot can be matched with the reference spot (L) to make a significant improvement in a number recognition rate, and since the system can be built without the addition of extra components or devices, costs for installation and management can be saved. Moreover, considering a feature that number recognition within a photographing area (S1) becomes optimal depending on neighboring conditions such as illuminance, time, fog and the like, when an image is obtained through a camera turned on by a controller, the image is divided into a plurality of pieces in a heightwise direction corresponding to a road moving direction, and then, the resolution of each divided pixel section is detected to reset a divided pixel section with the highest resolution as the position of the reference spot (L), so the position of the reference spot (L) is optimally changed in response to a neighboring condition to make the number recognition more accurate.

Description

Photographing system and method for increasing recognition rate of vehicle number}

The present invention relates to a vehicle photographing system and method for increasing the license plate recognition rate, and more particularly, by matching the point where the license plate of the intermittent vehicle is photographed with the preset reference point which is the point where the license plate is optimally recognized. It relates to a vehicle shooting system and method that can maximize the number recognition rate compared to the configuration and environment.

As the automobile industry developed and roads expanded, vehicle penetration increased, and the number of traffic law violations such as speed, signals, and reverse driving also increased exponentially in proportion to the increase in vehicle penetration.

These violations of traffic laws can lead to not only one's own life but also other people's lives, and various studies are being conducted on crackdown equipment and crackdown systems to crack down on them.

In particular, the camera of the crackdown equipment is widely applied to various crackdown equipment due to the identification and evidence function of the crackdown vehicle and the advantages that can be used for traffic information monitoring.

FIG. 1 is an exemplary view showing a conventional vehicle photographing system, and FIGS. 2A, 2B, and 2C are exemplary views for explaining the problem of FIG. 1.

The conventional vehicle photographing system (hereinafter, referred to as a prior art) 200 of FIG. 1 includes vehicle detecting means 201 for detecting an entering vehicle in the sensing area S, and sensing data detected by the vehicle detecting means 201. The controller 203 and the controller 203 which detects the intermittent vehicle and transmits a trigger signal to the camera 210 by predicting when the intermittent vehicle enters the photographing area S1 when detecting the intermittent vehicle. When the trigger signal is received from 203, the camera 210 performs a photographing operation to acquire an image of the intermittent vehicle.

In addition, the camera 210 is installed to capture a predetermined shooting area (S1). In this case, as the focus of the camera 210 is set toward the reference line L, the license plate recognition can be made more accurate as the photographing point of the intermittent vehicle C is closer to the reference line L.

However, the conventional technology 200 is the camera 210 because the controller 203 is configured to transmit the trigger signal after a predetermined time from the time when the interrupted vehicle C is detected regardless of the speed and the detection position of the intermittent vehicle. Even if the image of the intermittent vehicle C is obtained by the above, the image is taken when the license plate of the intermittent vehicle C is spaced apart from the reference line L, as shown in FIGS. 2A, 2B, and 2C. As a result, a phenomenon in which number recognition is not normally performed is occurring randomly.

In other words, there is an urgent need for research to match the license plate photographing point, which is the point where the license plate of the intermittent vehicle is photographed, with the preset reference line, which is the point where the license plate recognition is optimal.

In Korean Patent No. 10-1372838 (name of the invention: a multifunctional vehicle photographing system and method), the sharpness of these frames is determined by comparing the sharpness of frames of an image obtained by a camera, and the determined representative frame is determined. Disclosed is a multifunctional vehicle photographing system capable of improving number recognition rate by performing number recognition.

However, in the multi-function vehicle photographing system, when an image is obtained by the vehicle photographing apparatus, operations such as a calculation process for calculating the sharpness of each of the frames of the acquired image and an operation process for comparing the sharpness of each frame are essentially performed. Because of this, the operation process has a complicated problem.

In addition, the multi-function vehicle photographing system does not describe a separate technology and method for matching the photographing point of the license plate to the reference line (L) at all, so that the frames of the acquired image are not photographed at the optimal focus, and thus the license plate recognition rate. There is a structural limitation that does not solve the problem of this deteriorating prior art.

The present invention is to solve such a problem, the problem of the present invention is to use the location and speed of the intermittent vehicle, and the position of the predetermined reference point (L) where the vehicle number recognition is optimal when the intermittent vehicle is detected. After calculating the elapsed time (t) from the current position to the reference point (L) from the intermittent vehicle, the trigger signal is generated so that the camera is photographed at the calculated elapsed time (t). It is to provide a vehicle photographing system and method that can maximize the number recognition rate compared to the same environment and configuration, without adding a separate device and configuration means.

In addition, another object of the present invention is to take into account the characteristic that the pixel area where the number recognition is optimal in the shooting area (S1) is changed according to the surrounding environment such as illumination, time, fog, etc., the controller drives the camera to acquire an image Then, after dividing the obtained image into a plurality of height directions corresponding to the road movement direction, the sharpness of each of the divided pixel sections is detected and the position of the reference point L is reset to the divided pixel section having the highest sharpness. It is to provide a vehicle photographing system and method that can be more precisely number recognition by changing the position of the reference point (L) in response to the environment.

The solution means of the present invention for solving the above problem is a vehicle detecting means for detecting the entry vehicle of the preset detection area (S), a camera for photographing the preset shooting area (S1), the vehicle detection means and the camera In the vehicle photographing system comprising a controller for managing and controlling the controller: The controller analyzes the detection data detected by the vehicle detection means to detect the entry vehicle in the detection area (S), the position of the detected entry vehicle And a vehicle information generator for generating vehicle information including a speed; An intermittent vehicle determination unit that determines whether the entry vehicle is an interrupted vehicle by comparing the vehicle information generated by the vehicle information generator with preset violation information; When the intermittent vehicle is detected by the intermittent vehicle discrimination unit, the intermittent vehicle uses the current position and speed of the intermittent vehicle and the position of the preset reference point L at which the number recognition is optimally performed. An elapsed time calculating unit that calculates an elapsed time t that is a time until reaching L); A trigger signal generator configured to generate a trigger signal so that the camera photographs the elapsed time t calculated by the elapsed time calculator; A controller for outputting a trigger signal generated by the trigger signal generator to the camera; And a number recognizing unit for recognizing a vehicle number by analyzing the image photographed by the camera, wherein the controller further includes a reference point setting unit for resetting the reference point every predetermined period T. Each time T), the camera is driven to input an image obtained by photographing the camera to the reference point setting unit, and the reference point setting unit uses the pixel position information of each of the predetermined divided pixel sections to input the image. An image dividing module for dividing the divided pixel sections; A sharpness detection module for detecting the sharpness of each of the divided pixel sections divided by the image segmentation module using a preset sharpness detection algorithm; A sharpness comparison module for comparing the sharpness of each divided pixel section detected by the sharpness detection module; And a reference point determination module configured to reset the reference point L to be included in the divided pixel section having the highest clarity after detecting the divided pixel section having the highest clarity in the sharpness comparison module. The elapsed time t is calculated by using the reference point L determined by the reference point determination module.

In addition, in the present invention, the elapsed time calculating unit preferably calculates the elapsed time (t) through the equation (1).

[Equation 1]

Where X is the position of the reference point (L), X1 is the position of the interrupted vehicle, and V is the speed of the interrupted vehicle.

delete

In the present invention, the divided pixel sections have a short length in the height direction of the image, are formed in a rectangular shape having a long length in the width direction, and are preferably stacked in the height direction of the image.

Another solution of the present invention is a vehicle photographing method for recognizing a vehicle number from an image of an intermittent vehicle. The vehicle photographing method includes a controller analyzing a sensed data detected by a vehicle detecting means to locate a vehicle. A vehicle information generating step of generating vehicle information including a speed; An intermittent vehicle determination step of determining, by the controller, whether the detected vehicle is an interrupted vehicle by comparing the vehicle information generated by the vehicle information generating step with preset violation information; The controller proceeds when it is determined that the sensing vehicle is the intervening vehicle in the step of determining the intermittent vehicle, and utilizes the current position and speed of the intermittent vehicle and the position of the preset reference point L where the number recognition is optimal. An elapsed time calculating step of calculating an elapsed time (t) which is a time from reaching this point to the reference point (L); A trigger signal generation step of generating, by the controller, a trigger signal so that the camera photographs the elapsed time t calculated by the elapsed time calculating step; A trigger signal output step in which the controller outputs a trigger signal generated by the trigger signal generation step to a camera, and the camera performs shooting; An image input step in which the controller receives the image acquired by the trigger signal output step from the camera; The controller may include a number recognition step of recognizing a vehicle number by analyzing the image input through the image input step, and the vehicle photographing method further includes a reference point setting step that is performed every preset period T. The reference point setting step may include: an image acquiring step, in which the camera acquires an image by performing photographing every period (T); An image input step in which the controller receives an image acquired by the image acquisition step; An image dividing step of the controller dividing an image input through the image input step into the divided pixel sections by using pixel position information of each of the divided pixel sections; A sharpness detecting step of detecting, by the controller, the sharpness of each of the divided pixel sections divided by the image dividing step by using a predetermined sharpness detecting algorithm; A sharpness comparison step of the controller comparing the sharpness of each divided pixel section detected by the sharpness detection step; And a reference point determining step of resetting the reference point L to be included in the divided pixel section having the highest clarity after detecting the divided pixel section having the highest sharpness in the sharpness comparison step, and calculating the elapsed time. Calculates an elapsed time t by using the reference point L determined by the reference point determining step, and the divided pixel sections have a short length in the height direction of the image, but have a long length in the width direction. It is formed to be formed to be stacked in the height direction of the image.

In addition, in the present invention, the elapsed time calculating step is to calculate the elapsed time (t) through the equation (1).

[Equation 1]

Where X is the position of the reference point (L), X1 is the position of the interrupted vehicle, and V is the speed of the interrupted vehicle.

delete

delete

delete

According to the present invention having the above-mentioned problems and solutions, the vehicle is configured to be accurately photographed at the time when the intermittent vehicle passes the reference point L by utilizing the current position and speed of the intermittent vehicle and the position of the reference point L. It is possible to remarkably improve the number recognition rate by matching the point where the license plate is photographed to the reference point (L).

In addition, according to the present invention can be implemented without adding a separate configuration means and apparatus can reduce the installation and operating costs.

In addition, according to the present invention in consideration of the characteristic that the pixel area where the number recognition is optimal in the photographing area S1 is changed according to the surrounding environment such as illumination, time, fog, etc., the controller drives the camera to acquire an image. The image is divided into a plurality of heights corresponding to the road movement direction, and then the sharpness of each of the divided pixel sections is detected and the position of the reference point L is reset to the divided pixel section having the highest sharpness to respond to the surrounding environment. By optimally changing the position of the reference point (L) it is possible to more accurately recognize the number.

1 is an exemplary view showing a conventional vehicle photographing system.
(A), (b), (c) of FIG. 2 is an illustration for explaining the problem of FIG.
3 is a block diagram showing a vehicle photographing system according to an embodiment of the present invention.
4 is an exemplary view of FIG. 3.
5 is a block diagram illustrating the controller of FIG. 3.
6 is a flowchart for explaining the operation of the present invention.
(A), (b), (c) and (d) of FIG. 7 are exemplary views showing images obtained by the camera of the present invention.
8 is a block diagram illustrating a reference point setting unit of FIG. 5.
9A, 9B, and 9C are exemplary diagrams for describing the reference point setting unit of FIG. 5.
FIG. 10 is an exemplary diagram for describing divided pixel sections applied to FIG. 8.
FIG. 11 is a flowchart for describing an operation of the reference point setting module of FIG. 10.

Hereinafter, with reference to the accompanying drawings will be described an embodiment of the present invention.

3 is a block diagram showing a vehicle photographing system according to an embodiment of the present invention, Figure 4 is an exemplary view of FIG.

The vehicle photographing system 1 according to an embodiment of the present invention matches the photographing point of the license plate with a preset photographing point L where number recognition is optimally performed, without having to add a separate configuration means. This is to maximize the recognition rate.

3 and 4, the vehicle photographing system 1 receives a radar signal that is reflected after transmitting the radar signal and detects a vehicle entering the preset detection area S by the radar transceiver 4. Also, radar signals transmitted and received by the radar transceiver 4 are analyzed to generate vehicle information and detect an interrupted vehicle, and the interrupted vehicle arrives at a preset reference point L using the current position and speed of the interrupted vehicle. An image received from the camera 5 by outputting a trigger signal to the camera 5 so as to calculate the elapsed time t up to the following and to photograph the camera 5 described later at the elapsed time t. The controller 3-1, ..., (3-N) that recognizes the vehicle number by analyzing the Shoot the set shooting area (S1) to obtain the image of the intermittent vehicle and Traffic control server for receiving and storing vehicle information, control information and images from the camera 5 and the controllers 3-1, ..., (3-N) to the controller 3 ( 7) and a communication network 10 that provides a data movement path between the traffic control server 7 and the controller 3.

In the present invention, for the convenience of description, the vehicle detecting means for detecting the vehicle passing through the preset detection area S is described as an example of the radar transceiver 4, but the vehicle detecting means is not limited thereto. It is obvious that various known sensing means can be applied.

In addition, although not shown in Figure 2, the controller (3-1), ..., (3-N) is connected to the camera 5 and the radar transceiver 4 assigned to them by wired or wireless communication network (not shown) wired and wireless And transmit and receive data with each other via a communication network.

In addition, in FIGS. 2 and 3, only a single camera 5 is connected to each controller 3 for convenience of explanation. For example, the controller 3 is configured to manage and control the plurality of cameras 5. Can be.

The communication network 10 provides a data movement path between the traffic control server 7 and the controllers 3-1, ..., (3-N), and in detail, constitutes a wide area network (WAN) or the like. Can be.

The radar transceiver 4 transmits the radar to the preset sensing area S and receives the reflected signal.

In addition, the radar transceiver 4 periodically transmits the transmitted and received radar signal information to the controller 3.

The camera 5 is installed in the gantry 900 which is installed perpendicular to the moving direction of the road, and is installed to photograph the preset photographing area S1.

In addition, when the camera 5 receives a trigger signal from the controller 3, the camera 5 acquires an image by photographing and transmits the acquired image to the controller 3.

In this case, the photographing area S1 of the camera is formed in front of the sensing area S of the radar transceiver 4.

The controllers 3-1, ..., (3-N) control and manage the radar transceiver 4 and the camera 5 assigned to them.

In addition, the controller 3 analyzes the radar signal received from the radar transceiver 4 to detect the entry vehicle C in the detection area S, and calculates the position and speed of the entry vehicle to generate vehicle information. At this time, the vehicle information is composed of the position, time, speed, etc. of the sensing vehicle, and the vehicle information is used as traffic information in the traffic control server 7.

In addition, the controller 3 compares the generated vehicle information with preset violation information to determine whether the detected vehicle violates the law.

In addition, when the sensing vehicle is determined to be an interrupted vehicle, the controller 3 utilizes the position X1 and the speed of the interrupted vehicle and the position X of the preset reference point L, and the interrupted vehicle C becomes the reference point ( The elapsed time t until reaching L) is calculated, and a trigger signal is sent to the camera 5 in accordance with the elapsed time t. In this case, since the controller 3 calculates the elapsed time t based on the tip of the crackdown vehicle C, the shooting point of the license plate disposed at the tip of the crackdown vehicle C can be matched with the reference point L as much as possible. It becomes possible.

At this time, the reference point (L) is defined as the point where the number recognition in the photographing area (S1) is most accurate, and consists of a line having a width smaller than the width direction of the road perpendicular to the direction of movement of the road.

That is, the controller 3 takes the camera 5 into consideration when the license plate of the intermittent vehicle C passes the reference point L in consideration of the current position X1 and the speed in the sensing area S of the intermittent vehicle C. ) Is configured to take a picture, it is possible to maximize the number recognition rate compared to the same configuration and environment, without adding a separate configuration means and equipment.

In addition, when the controller 3 receives the image acquired from the camera 5, the controller 3 analyzes the received image using a preset number recognition algorithm to detect the vehicle number. At this time, since the number recognition algorithm is a technique commonly used in the control system and the parking control system, a detailed description thereof will be omitted.

In addition, when the vehicle number of the crackdown vehicle is detected, the controller 3 generates crackdown information that matches the position, time, violation, speed, vehicle number, and image frame of the crackdown vehicle, and generates the crackdown information using the traffic control server ( 7) to transmit. At this time, the traffic control server 7 performs a subsequent procedure such as a penalty charge for the crackdown vehicle using the crackdown information received from the controller 3.

The traffic control server 7 is a server that manages and controls the controllers 3-1, ..., (3-N), and transmits them from the controllers 3-1, ..., (3-N). Traffic information is generated by processing and utilizing the received vehicle information, and the generated traffic information is provided to the connected clients (not shown).

In addition, when the traffic control server 7 receives the enforcement information from the controllers 3-1, ..., (3-N), the traffic control server 7 performs a subsequent procedure such as a fine charge according to the received enforcement information.

5 is a block diagram illustrating the controller of FIG. 3.

As shown in FIG. 5, the controller 3 includes a controller 30, a memory 31, a communication interface 32, a signal analyzer 33, a vehicle information generator 34, and an intermittent vehicle determination unit. (35), an elapsed time calculating section 36, a trigger signal generating section 37, a number recognition section 38, an intermittent information generating section 39, and a reference point L setting section 40.

The control unit 30 is an operating system (O, S) of the controller 3, the control target 31, (32), (33), (34), (35), (36), (37), ( Manage and control 38), (39), (40).

In addition, when the control unit 30 receives the radar signal information from the radar transceiver 4, the control unit 30 inputs the received radar signal information to the signal analysis unit 33.

In addition, when the control vehicle 30 detects the control vehicle by the control vehicle determination unit 35, the control unit 30 executes the elapsed time calculation unit 36.

In addition, when the trigger signal (Trigger Signal) is generated by the trigger signal generation unit 37, the control unit 30 transmits the generated trigger signal to the camera (5).

In addition, when the controller 30 receives an image from the camera 5, the controller 30 inputs the received image to the number recognition unit 38.

In addition, the control unit 30 controls the communication interface unit 32 so that the vehicle number generated by the vehicle information generation unit 34 and the enforcement information generated by the enforcement information generation unit 39 are transmitted to the traffic control server 7. To control.

In addition, the controller 30 executes the reference point L setting unit 40 according to a preset period T, and simultaneously executes the trigger signal generator 37 to refer to the image photographed by the camera 5. Input to the point setting unit 40.

The memory 31 stores position information of the preset sensing regions S, position information of the preset photographing region S1, and position information of the reference point set by the reference point setting unit 40.

In addition, the memory 31 temporarily stores vehicle information generated by the vehicle information generation unit 34 and enforcement information generated by the enforcement information information generation unit 38.

In addition, the memory 31 stores a preset number recognition algorithm.

The communication interface 32 transmits and receives data with the traffic control server 7.

The signal analysis unit 33 receives the radar signal transmitted and received by the radar transceiver 4 and analyzes it.

The vehicle information generator 34 generates vehicle information by using the analysis data detected by the signal analyzer 33. At this time, the vehicle information consists of the position, time, speed, etc. of the sensing vehicle.

At this time, the vehicle information generated by the vehicle information generation unit 34 is transmitted to the traffic control server 7 through the communication interface unit 32 under the control of the control unit 30.

The enforcement vehicle determination unit 35 compares the vehicle information of the detection vehicle generated by the vehicle information generation unit 34 with a preset reference value to determine whether the detection vehicle violated the traffic laws, and when it is determined that the violation of the laws and regulations, The sensing vehicle is determined as an intermittent vehicle.

For example, the intermittent vehicle determining unit 35 may be configured to determine the detected vehicle as the intervening vehicle when the speed of the sensing vehicle exceeds the speed limit by comparing the speed of the sensing vehicle with the speed limit.

In addition, when the detection vehicle is determined to be the enforcement vehicle by the enforcement vehicle determination unit 35, the elapsed time calculation unit 36 is executed under the control of the control unit 30.

The elapsed time calculation unit 36 is executed when the detection vehicle is determined to be the control vehicle by the control vehicle determination unit 35.

In addition, the elapsed time calculation unit 36 calculates the separation distance Δd by utilizing the current position X1 of the intermittent vehicle and the position X of the reference point L set by the reference point setting unit 40. The elapsed time t from the current position to the reference point L is calculated by dividing the speed of the intermittent vehicle by the calculated separation distance Δd. It is defined by Equation 1.

X is the position of the reference point (L), X1 is the position of the intermittent vehicle, V is the speed of the intermittent vehicle.

In addition, the controller 30 controls the trigger signal generator 37 to correspond to the elapsed time t so that the camera 5 photographs the elapsed time t calculated by the elapsed time calculator 36. Run it.

When the trigger signal generator 37 is executed under the control of the controller 30, the trigger signal generator 37 generates a trigger signal for driving the camera 5, and the generated trigger signal is controlled by the controller 30. Is output.

The number recognition unit 38 receives an image of an intermittent vehicle obtained by photographing the camera 5, and analyzes the input image using a preset number recognition algorithm to recognize and detect a vehicle number. In this case, after extracting a license plate from an image or an image, a method and technology for recognizing a license plate number of the extracted license plate are commonly used techniques, and thus detailed description thereof will be omitted.

In addition, the vehicle number information detected by the number recognition unit 38 is input to the enforcement information generation unit 39.

The enforcement information generation unit 39 generates the enforcement information by matching the position, time, violation details, speed, vehicle number, and image of the enforcement vehicle.

At this time, the enforcement information generated by the enforcement information generation unit 39 is transmitted to the traffic control server 7 through the communication interface unit 32 under the control of the control unit 30.

6 is a flowchart for explaining the operation of the present invention.

As shown in FIG. 6, the vehicle photographing method S1 of the present invention includes a radar signal transmission and reception step S10, a vehicle detection determination step S20, a vehicle information generation step S30, and a vehicle information transmission step S40. , Intermittent vehicle determination step (S50), elapsed time calculation step (S60), trigger signal generation step (S70), trigger signal output step (S80), shooting step (S90), image input step (S100), number recognition step ( S110, intermittent information generating step (S120), intermittent information transmission step (S130).

Radar signal transmission and reception step (S10) is a step in which the radar transceiver 4 collects the reflected signal after transmitting the radar signal.

The vehicle detection determination step S20 is a step in which the controller 3 analyzes the radar signals transmitted and received by the radar signal transmission and reception step S10 to determine whether an entry vehicle exists in the preset detection area S.

In addition, if the vehicle detection step (S20) is not detected, the vehicle does not perform a separate operation, if the vehicle is detected, the vehicle information generation step (S30) proceeds.

The vehicle information generation step S30 is performed when the vehicle is detected by the controller 3 in the vehicle detection determination step S20. The vehicle information generation step S30 is a step of generating vehicle information including position, time, and speed of the vehicle.

The vehicle information transmitting step S40 is a step in which the controller 3 transmits the vehicle information generated by the vehicle information generating step S30 to the traffic control server 7.

The intermittent vehicle determination step S50 determines whether the detection vehicle violates the law by comparing the vehicle information generated by the vehicle information generation step S30 with a preset reference value.

In addition, the intermittent vehicle determination step S50 returns to the vehicle detection step S20 and repeats the process after the detection vehicle is not in violation of the law.

In addition, if the detection vehicle is a vehicle in violation of the law, the crackdown vehicle determination step (S50) determines the detected vehicle as the crackdown vehicle, and then proceeds to the elapsed time calculation step (S60).

Elapsed time calculation step (S60) proceeds when the controller 3 is determined by the control vehicle determination step (S50) to determine the vehicle is an intermittent vehicle, the current position and the reference point set by the reference point setting unit 40 of the intermittent vehicle The distance (Δd) of the point (L) is calculated, and the elapsed time (t) from the current position to the reference point (L) is obtained by dividing the speed of the enforcement vehicle by the calculated distance (Δd). It is a step of calculating.

The trigger signal generation step S70 is a step in which the controller 3 generates a trigger signal for driving the camera 5 at a time point corresponding to the elapsed time t calculated in the elapsed time calculation step S60.

The trigger signal output step S80 is a step in which the controller 3 outputs the trigger signal generated by the trigger signal generation step S70 to the camera 5.

In the photographing step S90, the camera 5 acquires an image of the intermittent vehicle by photographing according to the trigger signal received through the trigger signal output step S80.

The image input step S100 is a step in which the controller 3 receives an image obtained by the photographing step S90 from the camera 5.

The number recognition step S110 is a step in which the controller 3 recognizes the vehicle number by analyzing the image input through the image input step S100 using a preset number recognition algorithm.

In the control information generation step S120, the controller 3 recognizes the vehicle number recognized by the number recognition step S110, the image input by the image input step S100, and the position, time, speed, and violation of the control vehicle. In this step, the content is matched to generate crackdown information.

The enforcement information transmission step S130 is a step in which the controller 3 transmits the enforcement information generated by the enforcement information generation step S120 to the traffic control server 7.

(A), (b), (c) and (d) of FIG. 7 are exemplary views showing images obtained by the camera of the present invention.

Looking at the image 800 obtained by the camera 5 of the present invention with reference to (a), (b), (c), (d) of Figure 7, the point of the license plate of the intermittent vehicle (C) and It can be seen that the reference point (L) is matched, thereby increasing the number recognition rate in the controller (3).

FIG. 8 is a block diagram illustrating a reference point setting unit of FIG. 5, and FIGS. 9A, 9B and 9C are exemplary views for explaining the reference point setting unit of FIG. 5.

The reference point setting unit 40 is executed under the control of the controller 30 at every preset period T, and optimally resets the position of the reference point L in response to the surrounding environment.

In general, the pixel area in which the number recognition is optimal in the photographing area S1 has a characteristic that varies depending on the surrounding environment such as illuminance, time, and fog.

For example, the image 700 of the photographing area S1 may have the sharpest pixel area in the image 700 when the surrounding environment is 'A' as shown in FIG. 9A according to the surrounding environment. S1 ′ may be formed at an upper portion, and as shown in FIG. 9B, when the surrounding environment is 'B', the pixel region S1 ′ having the highest sharpness in the image 700 is lowered. As illustrated in (c) of FIG. 9, when the surrounding environment is 'C', the pixel area S1 having the highest definition in the image 700 may be formed in the middle.

In the present invention, the reference point is reset according to a predetermined period T through the reference point setting unit 40 in consideration of such characteristics.

As shown in FIG. 8, the reference point setting unit 40 includes an image input module 401, an image segmentation module 402, a sharpness detection module 403, a sharpness comparison module 404, and a reference point determination module ( 405).

In this case, the reference point setting unit 40 is executed every predetermined period T, and receives the image obtained by photographing by the camera 5 under the control of the controller 30 at the time of execution.

The image input module 401 receives an image obtained by photographing the camera 5.

FIG. 10 is an exemplary diagram for describing divided pixel sections applied to FIG. 8.

The image dividing module 402 inputs the image 700 input by the image input module 401 by utilizing pixel position information of each of the preset divided pixel sections 710-1,..., And 710 -N. Are divided into divided pixel sections 710-1, ..., (710-N).

In this case, each of the divided pixel sections 710-1,..., 710 -N has a long length in the width direction of the image 700, but has a short length in the height direction of the image 700. Is formed.

In addition, the divided pixel sections 710-1,..., And 710 -N are formed to be stacked on each other in the height direction of the image 700.

The sharpness detection module 403 detects the sharpness of each of the divided pixel sections 710-1,..., 710 -N divided by the image segmentation module 402 by using a preset sharpness detection algorithm. . In this case, since an algorithm for detecting the sharpness of a specific section from the image is a commonly used technique, a detailed description thereof will be omitted.

The sharpness comparison module 404 compares the sharpnesses of the divided pixel sections 710 detected by the sharpness detection module 403.

The reference point determination module 405 detects the divided pixel section having the highest sharpness through the sharpness comparison module 404, and then determines the reference point L to be included in the detected pixel section having the highest sharpness. For example, the reference point determination module 405 may set the lower end of the divided pixel section having the highest sharpness as the reference point L.

In addition, the reference point L determined by the reference point determination module 405 is used to calculate the elapsed time in the elapsed time calculation unit 36 until the next period T.

FIG. 11 is a flowchart for describing an operation of the reference point setting module of FIG. 10.

Reference point setting step (S400) of the present invention proceeds for each predetermined period (T).

In addition, as shown in FIG. 11, the reference point setting step S400 includes an image acquisition step S410 in which the camera 5 acquires an image by capturing every period T, and the controller acquires an image in step S410. Image input step (S420) of receiving the image obtained by the second), and the controller 5 utilizes pixel position information of each of the preset pixel divisions 710-1, ..., 710-N. Image division step (S430) of dividing the image input through the image input step (S420) into the divided pixel sections, and the controller 5 by the image segmentation step (S430) using a preset sharpness detection algorithm. Sharpness detection step (S440) for detecting the sharpness of each divided pixel section, and sharpness comparison step (S450) in which the controller 5 compares the sharpness of each divided pixel section detected by the sharpness detection step (S440). In the sharpness comparison step (S450), the sharpness is increased. After detecting the high divided pixel section, and the detected contrast it is to be included in the high split pixel section composed of a reference point of the reference point determining step (S460) for resetting the (L).

At this time, the elapsed time calculating step (S60) of FIG. 6 described above calculates the elapsed time (t) by using the reference point L determined by the reference point determining step (S460).

As described above, the vehicle photographing system 1 according to an exemplary embodiment of the present invention utilizes the current position and speed of the intervening vehicle C and the position of the reference point L so that the intervening vehicle C passes the reference point L. Since the photographing is accurately performed at the viewpoint, the number recognition rate can be remarkably improved by matching the photographing point of the license plate with the reference point (L).

In addition, since the vehicle imaging system 1 of the present invention can be implemented without adding a separate configuration means and apparatus, the installation and operation costs can be reduced.

In addition, the vehicle photographing system 1 of the present invention allows the controller to drive the camera in consideration of the characteristic that the pixel area where the number recognition is optimally performed in the photographing area S1 is changed according to the surrounding environment such as illumination, time, fog, and the like. After acquiring the image, the obtained image is divided into a plurality of heights in the height direction corresponding to the road moving direction, and then the sharpness of each of the divided pixel sections is detected to determine the position of the reference point L as the divided pixel section having the highest sharpness. By resetting, the position of the reference point L is optimally changed corresponding to the surrounding environment so that the number recognition can be made more accurately.

1: Vehicle shooting system 3-1, ..., 3-N: Controllers
4: Radar transceiver 5: Camera 7: Traffic control server
10: communication network 30: control unit 31: memory
32: communication interface unit 33: signal analysis unit 34: vehicle information generation unit
35: intermittent vehicle discrimination unit 36: elapsed time calculation unit
37: trigger signal generation unit 38: number recognition unit 39: enforcement information generation unit
401: image input module 402: image segmentation module
403: sharpness detection module 404: sharpness comparison module
405: reference point determination module

Claims (7)

A vehicle photographing system including vehicle sensing means for sensing an entrance vehicle in a preset sensing area S, a camera photographing a preset photographing area S1, and a controller for managing and controlling the vehicle sensing means and the camera. In:
The controller
A vehicle information generator configured to analyze the sensing data detected by the vehicle detecting means, detect an entry vehicle in the detection area S, and generate vehicle information including a position and a speed of the detected entry vehicle;
An intermittent vehicle determination unit that determines whether the entry vehicle is an interrupted vehicle by comparing the vehicle information generated by the vehicle information generator with preset violation information;
When the intermittent vehicle is detected by the intermittent vehicle discrimination unit, the intermittent vehicle uses the current position and speed of the intermittent vehicle and the position of the preset reference point L at which the number recognition is optimally performed. An elapsed time calculating unit that calculates an elapsed time t that is a time until reaching L);
A trigger signal generator configured to generate a trigger signal so that the camera photographs the elapsed time t calculated by the elapsed time calculator;
A controller for outputting a trigger signal generated by the trigger signal generator to the camera;
A number recognition unit for recognizing a vehicle number by analyzing the image photographed by the camera,
The controller further includes a reference point setting unit for resetting the reference point every predetermined period (T),
The controller drives the camera at each cycle T to input an image obtained by photographing the camera to the reference point setting unit,
The reference point setting unit
An image splitting module for dividing an input image into the divided pixel sections by using pixel position information of each of the divided pixel sections;
A sharpness detection module for detecting the sharpness of each of the divided pixel sections divided by the image segmentation module using a preset sharpness detection algorithm;
A sharpness comparison module for comparing the sharpness of each divided pixel section detected by the sharpness detection module;
And a reference point determination module for resetting the reference point L so as to be included in the divided pixel section having the highest clarity after detecting the divided pixel section having the highest clarity in the sharpness comparison module.
And the elapsed time calculating unit calculates the elapsed time (t) by using the reference point (L) determined by the reference point determining module. The vehicle photographing system of claim 1, wherein the elapsed time calculating unit calculates an elapsed time t through Equation 1.
[Equation 1]

Where X is the position of the reference point (L), X1 is the position of the interrupted vehicle, and V is the speed of the interrupted vehicle. delete The vehicle photographing method of claim 2, wherein the divided pixel sections have a short length in the height direction of the image, are formed in a rectangular shape having a long length in the width direction, and are stacked in the height direction of the image. system. In a vehicle photographing method for recognizing a vehicle number from an image of an intermittent vehicle:
The vehicle shooting method
A vehicle information generating step of the controller analyzing the sensed data detected by the vehicle detecting means and generating vehicle information including the position and the speed of the sensing vehicle;
An intermittent vehicle determination step of determining, by the controller, whether the detected vehicle is an interrupted vehicle by comparing the vehicle information generated by the vehicle information generation step with preset violation information;
The controller proceeds when it is determined that the sensing vehicle is the intervening vehicle in the step of determining the intermittent vehicle, and utilizes the current position and speed of the intermittent vehicle and the position of the preset reference point L where the number recognition is optimal. An elapsed time calculating step of calculating an elapsed time (t) which is a time from reaching this point to the reference point (L);
A trigger signal generation step of generating, by the controller, a trigger signal such that the camera photographs the elapsed time t calculated by the elapsed time calculating step;
A trigger signal output step in which the controller outputs a trigger signal generated by the trigger signal generation step to a camera, and the camera performs shooting;
An image input step in which the controller receives the image acquired by the trigger signal output step from the camera;
The controller comprises a number recognition step of recognizing the vehicle number by analyzing the image input through the image input step,
The vehicle photographing method further includes a step of setting a reference point that is performed every predetermined period T.
The reference point setting step
An image acquiring step, wherein the camera acquires an image by performing photographing every period T;
An image input step in which the controller receives an image acquired by the image acquisition step;
An image dividing step of the controller dividing an image input through the image input step into the divided pixel sections by using pixel position information of each of the divided pixel sections;
A sharpness detecting step of detecting, by the controller, the sharpness of each of the divided pixel sections divided by the image dividing step by using a predetermined sharpness detecting algorithm;
A sharpness comparison step of the controller comparing the sharpness of each divided pixel section detected by the sharpness detection step;
And a reference point determining step of resetting the reference point L so as to be included in the divided pixel section having the highest clarity after detecting the divided pixel section having the highest sharpness in the sharpness comparison step.
The elapsed time calculating step calculates the elapsed time (t) by using the reference point (L) determined by the reference point determination step,
The divided pixel sections may have a short length in the height direction of the image, are formed in a rectangular shape having a long length in the width direction, and are stacked in the height direction of the image. The method according to claim 5, wherein the elapsed time calculating step calculates an elapsed time (t) through Equation (1).
[Equation 1]

Where X is the position of the reference point (L), X1 is the position of the interrupted vehicle, and V is the speed of the interrupted vehicle. delete

Images