KR101833418B1 - Camera Controlling Method of CCTV System for Recognizing Car Number and the CCTV System - Google Patents

Abstract

A camera control method of a CCTV system for recognizing a car number and a CCTV system thereof are disclosed. The CCTV system according to the present invention basically corresponds to the change of an optical external environment according to the change of the position of the sun with time, by changing the iris setting of the camera in advance even though the vehicle number is based on a recognized image. The camera control method of the present invention is a kind of post-correction. Since an optical state change is continuously tracked and corrected, a problem that the vehicle number is not suddenly recognized according to the lighting state of sunlight is fundamentally solved. The camera control method includes a photographing step; a calculating step; and a camera adjustment step.

Description

Technical Field [0001] The present invention relates to a camera control method of a CCTV system for recognizing a car number,

The present invention relates to a camera control method of a CCTV system for vehicle number recognition, which solves the problem that vehicle number recognition in a vehicle image becomes impossible due to backlight illumination by the sun or reflection of sunlight on a license plate.

On the road, police vehicles and stolen vehicles are also running, and vehicles under trafficking are also driving. In order to detect such a "transgression vehicle" by the police, etc., it is a CCTV system for crime prevention using a camera in which it is easy to visually check a criminal vehicle and directly detect whether the vehicle is a transgression vehicle by using a radio.

A CCTV system for security of a vehicle usually detects a vehicle passing through a road using a laser beam, a loop coil or various other sensors. When a vehicle is detected, the vehicle is photographed using the camera, The car number is recognized. The recognized vehicle number is provided to an external control center along with the image of the vehicle, thereby conducting a theft or an arrest vehicle. Such a crime prevention vehicle recognition system has an advantage that it can reduce the manpower and time required for the search without causing a road congestion caused by the speedy search.

Among the methods of detecting a vehicle, a loop coil method in which a loop coil is embedded in a road is widely used in that the detection range is wide and the detection error is small.

On the other hand, in the actual road situation, various problems arise in the process of photographing the vehicle and recognizing the vehicle number from the captured image. One of them is that the sun is backlighted, so the entire background between the vehicle number and the car number is too bright, or sunlight is reflected on the license plate due to the sun being behind the camera, It is difficult to recognize the car number. Since the position of the sun changes in units of time, the external optical condition surrounding the subject constantly changes with respect to the fixed camera.

To solve this problem, it is necessary to adjust the shutter and the aperture. On the other hand, most cameras have an automatic brightness control function, which automatically adjusts the shutter speed or the opening of the iris if the average brightness of the entire image is too high or too low. The same applies to cameras used in CCTV systems for vehicle security.

However, the aperture control in the automatic mode is not directly related to the number recognition but is based on the brightness of the entire image, so that it may actually be difficult to recognize the text and background of the vehicle. Therefore, since the sun's position changes with time, even if the iris is set to be controlled automatically, the image of the car number is not recognized suddenly at any moment.

[Related Technical Literature]

1. Korean Patent Registration No. 1368831 (Vehicle Number Identification System and Method) A method of controlling a camera using an average brightness in an image or car number area is disclosed.

An object of the present invention is to provide a camera control method of a CCTV system for car number recognition for solving the problem that vehicle number recognition in a vehicle image becomes impossible due to backlighting of the sun or reflection of sunlight on a license plate, .

Accordingly, the objective of the present invention is to temporarily obscure the problem of suddenly photographing an image in which the vehicle number is not recognized by continuously tracking and compensating for the change of the optical environment while the external illumination changes with time .

In order to achieve the above object, the present invention discloses a camera control method in a CCTV system for detecting a vehicle and photographing an image of the vehicle. The method of the present invention includes: a photographing step of sensing a vehicle traveling on the road and generating an image of the vehicle taken by the camera; Recognizing a plurality of number-pixel areas corresponding to the car number and a background-pixel area corresponding to the background of the car number among the license plate area of the image generated by the camera, A calculation step of calculating a brightness difference of a pixel area; And adjusting a shutter speed and an aperture opening degree of the camera such that the brightness difference is equal to or greater than the first reference value when the brightness difference is smaller than the first reference value.

According to an embodiment, when the brightness difference is smaller than the first reference value, if the average brightness of the image is larger than the second reference value, the camera opening is reduced by a predetermined unit, A diaphragm fine adjustment step of adjusting the parameter value such that the diaphragm opening is widened by the predetermined unit; An adjustment confirmation step of repeating the photographing step and the calculation step after the diaphragm fine adjustment step; And repeating the diaphragm fine adjustment step and the adjustment confirmation step until the brightness difference becomes equal to or greater than the first reference value.

Furthermore, the method of the present invention may further comprise: a history management step of performing the shooting step, the calculating step, and the camera adjusting step for one day and storing the result on a daily basis; Extracting a parameter value having a maximum brightness difference for each time of day from a result of performing the history management step and setting a parameter value for each time period; Further comprising a diaphragm macro adjustment step of, when the brightness difference is smaller than the first reference value, applying the parameter value for each time period before the diaphragm fine adjustment step and performing the adjustment confirmation step, Can be adjusted quickly.

The present invention also has the right to a CCTV system that carries out the above method. The CCTV system of the present invention includes a camera, a vehicle sensing unit, a number recognition unit, and a camera control unit.

When the vehicle sensing unit senses a vehicle running on the road using the vehicle sensing sensor, the camera generates an image of the vehicle sensed by the vehicle sensing unit.

The number recognition unit recognizes a plurality of number-pixel areas corresponding to the car number and a background-pixel area corresponding to the background of the car number in the license plate area of the image generated by the camera, Calculate the brightness difference between the brightness and the background-pixel area.

The camera controller adjusts the shutter speed and the aperture opening degree of the camera such that the brightness difference is equal to or greater than the first reference value when the brightness difference is smaller than the first reference value.

The CCTV system according to the present invention basically adjusts the iris setting of the camera on the basis that the vehicle number is recognized so that the brightness difference between the vehicle number and the background is adjusted in advance so as to be optimal even if the vehicle number is recognized, And corresponds to the change of the optical external environment according to the change of the position of the sun in advance. The camera control method of the present invention fundamentally solves the problem that the vehicle number is not suddenly recognized according to the lighting condition of the sunlight, because the camera control method keeps track of and changes the change of the optical situation.

Further, since the current vehicle number is recognized, the aperture correction must be adjusted very finely. Accordingly, since the brightness difference between the number and the background is not a certain level, it may take a long time for the correction. The present invention can quickly adjust the optimal aperture opening degree in the corresponding time zone by managing the individual adjustment history data.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram of a CCTV system according to the present invention,
2 is a block diagram of a CCTV system according to an embodiment of the present invention.
FIG. 3 is a view showing license plate regions among images photographed by the camera of the present invention,
4 is a flowchart illustrating a camera control method of a CCTV system according to an embodiment of the present invention.

BEST MODE FOR CARRYING OUT THE INVENTION The present invention will be described in more detail with reference to the drawings.

Referring to FIG. 1, the CCTV system for crime prevention of the present invention includes at least one camera 101 for photographing a road (or a lane), vehicle sensors 103 and 105 for detecting a vehicle, And an image controller 110 for processing and storing images of the camera 101.

The camera 101 of the present invention is installed on a road, and generates an image of a vehicle running on the road.

The camera 101 is attached to a supporting structure provided at a predetermined height from the road surface so as to face the running direction of the vehicle, and the image controller 110 is generally installed outside the road from the camera 101. As described below, the camera 101 may be normally installed to photograph a plurality of lanes at the same time. For example, in the case of eight round trip lanes, you can use only two cameras in each of the four lanes or four cameras in all two lanes. In addition, a camera may be provided for each lane. Therefore, the CCTV system of the present invention may have one camera or a plurality of cameras installed in each lane. On the other hand, there is a road on which a plurality of vehicles can run at the same time without lane separation, and one or more cameras 101 of the present invention may be installed.

The camera 101 operates according to various parameter values for its operation. For example, the user can control the operation of the camera 101 by setting and adjusting parameter values relating to color correction, iris opening, shutter speed, and the like. In the present invention, after initial setting, the camera control unit 207 controls the parameter value to be adjusted by the camera control unit 207, do.

The vehicle sensors 103 and 105 sense a vehicle traveling along the road together with the vehicle sensing unit 203 of the image controller 110. [ The vehicle sensors 103 and 105 may be any type of sensor capable of detecting that the vehicle has passed through, or may be a loop coil type sensor installed on the road and detecting the vehicle by electromagnetic induction, And a sensor for detecting the vehicle using a laser. These vehicle sensors 103 and 105 can detect the speed of the vehicle at the same time. 1 is an example in which first and second vehicle sensors 103 and 105 of a loop coil system are installed in the first lane and the second lane, respectively.

However, the vehicle sensors 103 and 105 may not be essential components of the present invention. For example, in the CCTV system of the present invention, the camera 101 can continuously photograph a road without a separate vehicle sensor, and can detect the running of the vehicle through image processing of the captured image.

The image controller 110 senses a vehicle running on the road using the vehicle sensors 103 and 105 and controls the camera 101 to shoot the vehicle according to the detection result to generate an image, Save and manage. If necessary, the image controller 110 may perform image processing such as overspeed interception and license plate recognition.

2, the image controller 110 includes a storage medium 201, a vehicle sensing unit 203, a number recognizing unit 205, and a camera control unit 207 to control the iris opening of the camera 101 So that the camera 101 can generate an image suitable for number recognition.

The storage medium 201 stores an image captured by the camera 101 and provided.

The vehicle sensing unit 203 senses a vehicle running on the road using the vehicle sensors 103 and 105. [ The vehicle sensing unit 203 provides the camera 101 with the sensing result read from the vehicle sensors 103 and 105 to control the camera 101 to generate an image of the sensed vehicle.

As described above, the vehicle sensors 103 and 105 and the vehicle sensing unit 203 may not be essential components of the present invention. However, if the vehicle sensors 103 and 105 and the vehicle sensing unit 203 are not provided, the camera 101 is set to provide the image controller 110 with an image of continuously photographing the road.

The number recognizing unit 205 recognizes the license plate area 30 from the image captured by the camera 101 and recognizes the vehicle number. At this time, the number recognition unit 205 identifies a plurality of number-pixel areas 31 corresponding to the vehicle number and a background-pixel area 33 corresponding to the background of the vehicle number in the license plate area 30 do. 3, seven number-pixel areas 31 corresponding to seven-digit car numbers are recognized, and a background-pixel area 33 corresponding to the background of the car number is recognized.

The number recognizing unit 205 calculates the brightness of the recognized number-pixel area and the brightness of the background-pixel area, respectively, and then calculates the brightness difference between the number-pixel area and the background-pixel area (Hereinafter referred to as " brightness difference between backgrounds ") to the camera control unit 207.

Here, the brightness of the number-pixel region is a plurality of numbers-the brightness representing the pixel region 31, and a plurality of number-pixel regions 31, which average the brightness of all the pixels belonging to the pixel region 31 , Or may be a 'representative pixel average brightness' that averages the brightness of some pixels belonging to the multiple number-pixel region 31. [ The brightness of the background-pixel region may also be the background-brightness of the pixel region 33, which is the average brightness of all pixels belonging to the background-pixel region 33 , And 'representative pixel average brightness' which averages the brightness of some pixels belonging to the background-pixel area 33. [ Assuming that the brightness value of the pixel is represented by a value between 0 and 255, the brightness of the number-pixel region and the brightness of the background-pixel region are also between 0 and 255.

The brightness difference between the number-pixel area and the background-pixel area, that is, the difference in brightness between the number and the background, can be determined by a conventionally known method, which will be described below.

In addition, the number recognizing unit 205 may obtain the average brightness of the whole image and provide it to the camera controller 207 in order to perform the camera adjusting step described below.

The camera control unit 207 controls parameter values for the operation of the camera 101. Items controlled by the camera control unit 207 vary widely, such as color correction, shutter speed, aperture setting, etc. In the present invention, the control of opening of the diaphragm will be mainly described. The present invention controls the opening state of the diaphragm so that the image photographed by the camera 101 becomes an image free from the problem of recognizing the car number of the number recognizing unit 205. [

The camera control unit 207 determines whether the brightness difference between the number and the background calculated by the number recognition unit 205 is smaller than the first reference value. The difference in brightness between the normal number and the background should be more than a certain value. The background of the license plate is a bright color (normally white) with a brightness value of about 170 or so, and the number is blue, green, and black. The number recognition algorithm of the number recognition unit 205 includes a method of recognizing the outline of the number-pixel area. If the brightness difference between the number and the background is not sufficient at this time, the number recognizing unit 205 can not recognize the outline of the number-pixel area or may erroneously recognize the number, so that the number recognition failure or the number recognition error may occur. Therefore, the difference in brightness between the number and the background must always differ by more than a certain amount.

If the difference in brightness between the number and the background is less than the first reference value, the camera controller 207 performs a camera adjustment step of adjusting the iris opening degree of the camera such that the brightness difference is equal to or greater than the first reference value. Here, the first reference value can be obtained experimentally based on the brightness of the number-pixel area and the brightness of the background-pixel area obtained through the repeated experiment. For example, when the brightness value of the pixel is represented by a value between 0 and 255, The first reference value can be set to approximately 100 or so.

Hereinafter, with reference to FIG. 4, the camera control method of the present invention will be described focusing on the operations of the camera 101, the number recognition unit 205, and the camera control unit 207. FIG.

<Vehicle image shooting: S401>

The camera 101 generates a vehicle image. 2, when the vehicle sensors 103 and 105 and the vehicle sensing unit 203 are provided, when the vehicle sensing unit 203 senses a vehicle traveling on the road using the vehicle sensors 103 and 105 , And provides a detection result to the camera 101 to control the camera 101 to generate an image of the vehicle sensed as being traveling. At this time, the camera 101 generates an image of the vehicle while maintaining the aperture opening according to the current parameter value.

<Vehicle number recognition: S403>

The number recognizing unit 205 recognizes the license plate area 30 from the image captured by the camera 101 and recognizes the vehicle number. At this time, the number recognition unit 205 distinguishes the plurality of number-pixel areas 31 corresponding to the vehicle number and the background-pixel area 33 corresponding to the background of the vehicle number from the license plate area 30 .

<Calculation of brightness difference between number and background: S405>

The number recognition unit 205 recognizing the number-pixel area 31 and the background-pixel area 33 calculates the brightness of the number-pixel area and the brightness of the background-pixel area, and calculates the difference.

&Lt; Determining whether the brightness difference between the number and the background is larger than the first reference value: S407>

The camera control unit 207 determines whether the brightness difference between the number and the background calculated by the number recognition unit 205 is smaller than the first reference value.

If the difference in brightness between the number and the background is greater than the first reference value, the current aperture opening is appropriate, so the camera control unit 207 returns to step S401 and repeats steps S401 to S407 for the next image. The reason why the steps S401 to S407 are repeated even when there is no problem in the current state is that the position of the sun changes with the lapse of time and the external optical conditions change accordingly.

If the difference in brightness between the number and the background is less than the first reference value, the current iris opening degree is not appropriate. Therefore, the camera controller 207 adjusts the iris opening degree of the camera by performing the following camera adjustment step.

&Lt; camera adjustment - fine aperture adjustment step; S409 to S413>

When the brightness difference is smaller than the first reference value, the camera control unit 207 enters the camera adjustment step.

First, the camera control unit 207 determines a parameter value relating to aperture opening based on the average brightness of the entire image or the average brightness of the number-pixel area and the background-pixel area provided by the number recognizing unit 205.

When the iris parameter is adjusted, the amount of light incident on the camera 101 is increased or decreased. At this time, the brightness of the relatively bright background-pixel region is not increased or decreased at a certain rate . For example, if the diaphragm is slightly opened (for example, the diaphragm is opened about 5% more), the brightness of the background-pixel area becomes abruptly brighter as compared with the brightness of the number- The difference becomes large. Conversely, for example, when the aperture is slightly closed (for example, the aperture is closed by about 5%), the brightness of the number-pixel region becomes drastically darker than the brightness of the background-pixel region, The difference becomes large. Therefore, the camera control unit 207 of the camera adjustment step adjusts the parameter value to be slightly larger or slightly smaller, so that the brightness difference is larger than the first reference value.

First, the camera controller 207 determines whether the average brightness of the number-pixel area and background-pixel area provided by the number recognizer 205 is greater than the second reference value. As described above, the average brightness of the entire image may be used instead of the average brightness of the number-pixel area and the background-pixel area. Here, the second reference value can also be obtained experimentally. For example, when the brightness value of the pixel is represented by a value between 0 and 255, the second reference value may be set to about 160, which is higher than the middle value. However, the second reference value may be set to a value that changes with time rather than using a fixed value. Therefore, the second reference value of the daytime period can be set to a value larger than the morning or evening time period (S409).

If the average brightness of the number-pixel area and background-pixel area is greater than the second reference value, the camera controller 207 closes the aperture by a predetermined amount (S411), and if the average brightness of the number- If it is smaller than the reference value 2, control for opening the iris by a predetermined amount is performed (S413).

At this time, it is preferable to set the predetermined size to the same size based on the parameter value. The opening of the diaphragm is gradually increased or decreased through steps S411 or S413. As in the previous example, you can control the parameter value to reduce or open the aperture by 1% to 5%.

<Camera adjustment - adjustment confirmation step; Repetition of S401 to S413>

If the iris is adjusted through the above step S413, the influence of the iris adjustment is reflected on the photographed image thereafter. As described above, the image controller 110 of the present invention continuously repeats steps S401 to S413 after step S413 until the brightness difference becomes equal to or greater than the first reference value. However, according to the change of the sun, the difference between the number and the background becomes smaller than the first reference value in step S407, and if the new iris setting is found while repeating steps S401 through S413, the setting can be maintained for a considerable time.

The camera control of the present invention is performed through the above method. As described above, the image controller 110 of the present invention repeats S401 to S407 and repeats S401 to S413 as necessary. This process prevents the situation where the position of the sun changes with the lapse of time, and thus the number of the captured image is not recognized due to the change of the external optical condition. Furthermore, when an ordinary camera automatically controls the iris based on the average brightness, there may occur a situation where an optical state is deteriorated from the viewpoint of number recognition, and an image which is suddenly unable to be recognized is photographed. However, according to the present invention, The entire process is processed on the premise of number recognition. In other words, there is no problem in recognition of the current number, but it is characterized in that it does not create an unexpected unrecognizable state by responding to the movement of the sun in advance.

Embodiment (Aperture Macro Adjustment Method Based on History Data)

The difference between the number and the background becomes smaller than the first reference value in step S407 according to the change of the sun. If the new iris setting is found while repeating steps S401 through S413, the setting can be maintained for a considerable time. However, steps S401 to S413 are repeated several times in order to find the new setting value. In the case of a road where a lot of vehicles do not travel, the adjustment time may be considerable.

In order to solve this problem more actively, the method of the present invention can use the following method.

<History management stage>

First, the process of storing the result of performing the shooting step, the calculating step, and the camera adjusting step in one day in S401 is repeated several days, and the camera control unit 207 stores the control result for several days after the first installation, .

<Step of setting iris parameter value by time zone>

The camera control unit 207 extracts a parameter value having a maximum brightness difference between the number and the background for each day of the day from the result of performing the history management step, and sets the parameter value as 'time period parameter value'.

<Aperture Macro Adjustment Step Between S407 and S409>

While the camera 101 is performing the objective function of the CCTV system of the present invention while photographing the vehicle image, if the brightness difference is smaller than the first reference value as a result of step S407 of FIG. 3, the camera adjustment step is entered. However, before performing the camera adjustment step shown in FIG. 3 (that is, before proceeding to step S409), the parameter values for each time period in the corresponding time zone are applied first. After applying the time-based parameter values, the normal camera adjustment step is performed.

Through the macro adjustment step, the camera control unit 207 can adjust the iris opening degree more quickly.

Example 2 - Calculation of brightness of number pixel region

In calculating the brightness of the number-pixel region and the brightness of the background-pixel region, the present invention proposes the following method. The brightness of the number-pixel area and the background-pixel area is considered to be an important technical element itself, but it is meaningful as a means of determining whether or not the number recognition is possible. Therefore, the method of calculating the brightness may not be strict.

The number recognizing unit 205 may set the position of a plurality of specific pixels for each number / character recognized as a car number. The pixels may be arranged along the centerline of the number or along the edge.

While the camera 101 is performing the objective function of the CCTV system of the present invention while shooting the image of the vehicle, after recognizing the number by performing step S403 of FIG. 3, the corresponding number (for example, '3' Pixels corresponding to a predetermined plurality of predetermined pixel positions are extracted from the number-pixel area 31. [ The camera control unit 207 can calculate the brightness of the number-pixel area by calculating the average brightness of the extracted pixels.

In addition, the centerline of the number-pixel area 31 may be extracted, and the average brightness of the pixels belonging to the centerline may be calculated as the brightness of the number-pixel area.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is clearly understood that the same is by way of illustration and example only and is not to be construed as limiting the scope of the invention as defined by the appended claims. It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the present invention.

Claims (8)

A camera control method of a CCTV system for sensing a vehicle in motion and capturing an image of the vehicle,
A photographing step of sensing a vehicle traveling on a road and generating an image of the vehicle taken by the camera;
Recognizing a plurality of number-pixel areas corresponding to the car number and a background-pixel area corresponding to the background of the car number in the license plate area of the image generated by the camera, A calculation step of calculating a brightness difference of a pixel area; And
And adjusting a shutter speed and an aperture opening degree of the camera such that the brightness difference is equal to or greater than the first reference value when the brightness difference is smaller than a first reference value,
Wherein the camera adjusting step comprises:
If the brightness difference is smaller than the first reference value, if the average brightness of the image is larger than the second reference value, the iris opening is reduced by a predetermined unit, and if the average brightness is smaller than the second reference value, Adjusting the parameter value so that the parameter value is widened by a unit;
An adjustment confirmation step of repeating the photographing step and the calculation step after the diaphragm fine adjustment step; And
And repeating the diaphragm fine adjustment step and the adjustment confirmation step until the brightness difference becomes equal to or greater than the first reference value.
delete The method according to claim 1,
A history management step of performing the shooting step, the calculating step, and the camera adjusting step for one day and storing the result on a day-by-day basis; And
Extracting a parameter value having a maximum brightness difference in each day of the day from a result of performing the history management step and setting a parameter value for each time period; And
Further comprising a stop macro adjustment step of, if the brightness difference is smaller than the first reference value, applying the parameter value for each time slot prior to the diaphragm fine adjustment step and performing the adjustment check step / RTI &gt;
The method according to claim 1,
Wherein,
Recognizing the number-pixel area as a corresponding number or character in the vehicle image; And
Extracting a plurality of pixels corresponding to a plurality of coordinates preset to the character or number from the plurality of number-pixel areas, and calculating an average brightness of the extracted plurality of pixels as brightness of the number-pixel area, And controlling the camera in the CCTV system.
A CCTV system for sensing a vehicle in motion and photographing an image of the vehicle,
A vehicle sensing unit that senses a vehicle traveling on the road using a vehicle sensing sensor;
A camera for generating an image of the vehicle sensed by the vehicle sensing unit;
Recognizing a plurality of number-pixel areas corresponding to the car number and a background-pixel area corresponding to the background of the car number among the license plate area of the image generated by the camera, A number recognition unit for calculating a brightness difference of a pixel area; And
And a camera controller for adjusting a shutter speed and an aperture opening degree of the camera such that the brightness difference is equal to or greater than the first reference value when the brightness difference is smaller than a first reference value,
The camera control unit,
If the brightness difference is smaller than the first reference value, if the average brightness of the image is larger than the second reference value, the iris opening is reduced by a predetermined unit, and if the average brightness is smaller than the second reference value, After the fine adjustment to adjust the parameter value so as to widen the unit, the process of recalculating and confirming the brightness difference in the image taken again by the camera is repeated until the brightness difference becomes equal to or greater than the first reference value Features CCTV system.
delete 6. The method of claim 5,
The camera control unit,
The process of adjusting the iris opening degree with respect to the image photographed by the camera is repeated for several days, and the result is stored in units of days. Then, parameter values having the maximum brightness difference are extracted for each time zone, Lt; / RTI &gt;
Wherein when the brightness difference is smaller than the first reference value, the parameter value for the time period is applied prior to the diaphragm fine adjustment.
6. The method of claim 5,
Wherein,
Recognizing the number-pixel area as a corresponding number or character in the vehicle image, extracting a plurality of pixels corresponding to a plurality of predetermined coordinates from the plurality of number-pixel areas to the character or number, Wherein the average brightness of the plurality of pixels is calculated as the brightness of the number-pixel region.

Images