KR101625538B1

KR101625538B1 - Car Number Recognition system

Info

Publication number: KR101625538B1
Application number: KR1020150100793A
Authority: KR
Inventors: 최태완; 설성욱; 권용구; 박희문
Original assignee: 비원이미지 주식회사
Priority date: 2015-07-16
Filing date: 2015-07-16
Publication date: 2016-06-13

Abstract

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a license plate recognition system and video transmission capable of crime prevention in the city center. The present invention relates to a security camera system capable of transmitting moving images obtained by continuously photographing an object in a plurality of frames, wherein the computer judges an image of an object that has passed through a lens in a road on a road, And a method of controlling the shutter when the subject of the license plate is a vehicle. In the present invention, real-time license plate recognition using a heuristic division algorithm for pattern recognition is quickly performed by photographing an object clearly, which is robust against illumination change and noise, irrespective of objects of objects during daytime and nighttime, We propose an integrated security system that can simultaneously transmit and record using streaming.

Description

{Car Number Recognition system} A multi-

The present invention relates to a security camera system capable of realizing a moving picture obtained by continuously photographing an object in a plurality of frames. The security camera system includes an urban security camera system function that can be installed on an alley or a back road of a city center, The present invention relates to a control method capable of performing a rapid license plate recognizing function and capable of performing distributed processing for the same.

The license plate recognition system has been developed for efficient management of complex traffic environment and is now used in many places. However, since it is greatly influenced by environmental changes such as lighting, noise, background change, and license plate damage, performance is proved only in a limited environment, and there are still many difficulties in real- In the conventional art, it is practically difficult to accurately determine an object at night. Therefore, in order to solve this problem, it is conceivable to provide a camera in which a shutter speed is set suitably for a photographing condition of a person and a camera in which a shutter speed is set suitably for a photographing of a vehicle, separately. However, There is a problem that it must be used. In order to overcome the conventional problems, there has been an effort to acquire images of a plurality of images by a combination of a diaphragm and a shutter, and to process the image by setting the background, foreground, object,

Korean Patent Laid-Open Publication No. 10-2014-0116266 discloses a contour extracting algorithm using a conventional Sobel mask pattern in a preprocessing process for extracting a license plate area more accurately, since a multiplication operation is performed in an arithmetic process, In order to solve the problem of poor recognition rate, the conventional noise canceling algorithm using erosion and expansion requires a window scan algorithm consisting of addition and subtraction only, a feature of license plate area and noise area, or various situations There is known a method of image preprocessing for recognition of a license plate and a method of designing a hardware block for implementing the same, which is configured to extract a candidate region of a license plate quickly with a high recognition rate using the noise elimination algorithm considered.

Korean Patent Laid-Open No. 10-2015-0067623 discloses a license plate recognition method including the following three steps. Specifically, in the first step, when the ambient illuminance is lower than the reference illuminance, the first camera is used to output a trigger signal according to the vehicle entry speed. In the second step, the trigger signal is transmitted to the first camera The second camera is used to capture the image frames of the entrance vehicle, and the shutter opening time corresponding to the trigger signal and the illuminance of the illumination unit are applied. In the third step, the license plate area is extracted from the image frames from the second camera, and the license plate information is read from the extracted license plate area.

European Patent No. EP 02535881 relates to a method and apparatus for monitoring compliance with road traffic regulations, including compliance with speed limits. The automatic system proposed herein reduces the probability of error in determining a violating vehicle, increases the length of the speed limit to monitor areas of hundreds of thousands of meters, and builds and maintains mounting supports of the speed limit monitoring device To reduce costs. To this end, data flows from a video camera and a radar are separately obtained and then compared to obtain a combined processing of signals from a radar and a panoramic video camera in which data on speed and coordinates are obtained with almost no error probability in discriminating a violating vehicle A new method is proposed. Devices incorporating the proposed method are known to include a radar having a signal processing module that allows the calculation of the speed and distance of all vehicles on a selected road segment, and a panoramic video camera.

In Korean Patent No. 10-1493009, there is provided a method of recognizing a car number of a front and rear car number recognition system, comprising the steps of: photographing a front license plate of a vehicle in which a first camera moves in a first direction; 1) receiving a first frame including the front license plate photographed by a camera, recognizing the front license plate from the received first frame, and transmitting a first recognition number corresponding to the recognition result to the management server, The second camera captures a rear plate of the vehicle moving in the first direction, and the first recognition module receives a second frame including the rear plate taken by the second camera, Recognizing the rear plate from the second frame and transmitting a second identification number corresponding to the recognition result to the management server; Photographing the front license plate of the vehicle moving in a second direction opposite to the first license plate, receiving a third frame including the front license plate photographed by the third camera, Recognizing the front license plate from the frame, and transmitting a third identification number corresponding to the recognition result to the management server.

Recently, in order to prevent incidents and accidents on public roads and to solve incident incidents by reading captured images in the event of an accident, CCTV cameras are installed in predetermined places such as cities, for example, System installation and utilization is also increasing. In order to prevent and solve accidents using high-resolution cameras, it is necessary to at least identify people and vehicles accurately from the images taken from the cameras. That is, it is simultaneously necessary to identify the impression wearing including the face of the photographed person, the license plate, and the like.

However, the installation environment of the security-related system is different from the one in which only one image is processed with one camera. There is a camera system only for license plate recognition, and a video camera system for crime prevention in the city is separately installed. This causes many problems such as the duplication of urban security cameras, the difficulty of separate budgeting and management, and the deterioration of the aesthetics of the city and the different administrative agencies.

Therefore, there is an increasing tendency to apply the intelligent traffic system to various situations and environment. According to the demand, the license plate recognition system has been developed for efficient management of complicated traffic environment and is now used in many places. However, the vehicle license plate recognition system is difficult to be used in real time because it can operate only in a limited environment because it is greatly influenced by environmental changes such as illumination, noise, background change, and license plate damage.

delete

Korean Patent Laid-Open Publication No. 10-2014-0116266 Korean Patent Publication No. 10-2015-0067623 European Patent No. EP02535881 Korean Patent No. 10-1493009

Currently, the integrated video surveillance system is composed of a single camera that processes only one video according to the purpose. In total, two cameras are used for high-resolution LPR (License Plate Recognition) and low-resolution video storage, adding video video transmission equipment. Such a video storage method has difficulty in searching for a target object. The still image storage method may be a sensor method, a loop method, a laser method, or the like, and a video can be acquired using a motionless motion detection method where the method is not suitable for installation. The disadvantage of this method is that there is no sensor that does not receive the help of the hardware, and therefore the vehicle image may be missed due to the lowering of the detection rate.

The present invention proposes an integrated crime prevention system which solves the difficulties of missing still images during storage and retrieval during moving picture storage, which is a problem of the above-described conventional technology. The system is implemented using a single camera capable of simultaneous high-resolution LPR and high-resolution video storage. This system does not need to use additional video transmission equipment and it enables video transmission with one camera installed. In addition, two lanes can be processed using the software and the advantage of obtaining a wider image by improving the resolution and image acquisition method. And, when detecting using the loop, it is possible to check the missing information through the video information of the missing vehicle. The F-LPR (1-Camera and 2-Way Fast LPR) algorithm used in this system is a high-speed processing program that can recognize and process vehicle numbers of two lane images using one camera.

In order to solve the above-mentioned problem of the separate installation of the moving picture camera, the present invention transmits video streaming to a server using a frame compression technique and a video transmission technique from one security camera, Provides a security camera system.
In order to accomplish the above object, the present invention provides a program for controlling the shutter so as to photograph the subject while changing the brightness of the subject by determining the presence or absence of an object, the shutter speed that substantially controls the time during which the image passes through the lens, There is provided a security camera system that continuously photographs a vehicle or a person on the road in a plurality of frames, including substantial hardware control means.

The present invention provides an integrated security camera system that includes an object detection unit that detects an object of the object, and captures the object using an electrical signal applied according to the type of the detected object. The object detection unit may be configured as a control unit that can acquire an image by a physical sensor or an image processing technique using a photographed image.

The present invention relates to a heuristic division algorithm for accurately recognizing a license plate used in a plurality of intelligent traffic systems, irrespective of the position and angle of a vehicle in an input image, robust against illumination change and noise, And provides a real-time license plate recognition system.
The present invention refers to an index table created by statistical data to perform an appropriate shutter speed control (Optimal Shutter Speed Control) for obtaining a clear image of a vehicle and a person, Real-time moving picture.

The effects of the integrated security camera system and the control method thereof according to the present invention are as follows. According to the present invention, there is an advantage in that it is possible to shoot an optimal image by adjusting the shutter speed using the brightness information in an environment where the object ambient illuminance on the road changes. That is, the present invention has an advantage that it can acquire a high-quality image which can be read out regardless of the type and illuminance of an object, compared with a dark image obtained at night in the conventional method. In addition, according to the present invention, it is possible to process the license plate recognition and moving picture storage which are not processed in the conventional method at high speed in real time, and obtain good images at the same time regardless of the moving speed of people and vehicle.

1 is a schematic diagram of the entire integrated security camera system
2 shows an image acquisition process corresponding to one step of Camera trigger Timing (T)
3 is a view showing a control method of the camera when entering a vehicle
Fig. 4 is a flowchart illustrating a method
5 is a block diagram showing the configuration of an integrated security camera system according to the present invention.
6 is a block diagram showing the overall operation of the integrated security camera system according to the present invention.

delete

The basic factors affecting the amount of light incident on an image sensor of a camera such as a CCD or CMOS image sensor include intensity of a light or a light source, degree of reflection of an object, degree of exposure of a lens, exposure time, sensitivity of an image sensor, And a signal-to-noise ratio (SNR), which is electronic noise. The intensity of the light and the degree of reflection of the object are factors that can not be controlled by external factors, and the sensitivity of the image sensor is an element that is practically impossible to control unless it is a manufacturer. Therefore, a general camera adjusts the exposure of the object using the aperture (aperture) and the shutter speed (exposure time) due to the variation of the focus due to the refraction of the light, so that the object is photographed without out-focusing. That is, even if the aperture and the shutter speed are differently set, it is possible to combine the aperture and the shutter speed to obtain the same exposure effect. For example, it is possible to open the aperture and speed up the shutter speed, and to arrange the aperture and slow the shutter speed.

delete

In the conventional security camera system, five frames are typically shot in one second, and shutter speeds are the same in each frame. Another method is to take a continuous shot at different shutter speeds of each frame to be photographed. The iris is automatically variable within a predetermined range, but it corresponds to the brightness of the entire camera system installation site such as daytime and nighttime, and it is impossible to adjust in a short time instant compared with the brightness of the object.

In the prior art, both a person and a vehicle are photographed at night with a single shutter speed, which is practically applied to all frames in the same manner, and a method of photographing several frames successively by varying the shutter speed of each frame I tried to get the image. However, the image obtained by this method is a process of fixing the shutter speed or obtaining three frames at equal intervals of the shutter speed within the reference time. This is pointed out as a disadvantage because the image frame is missing when the vehicle speed is very high or there are many vehicles passing by. It is also true that the quality of the image due to the difference in brightness depending on the shutter speed is poor at the time of nighttime, the intersection with the nighttime at night, the sunrise in the morning, and the sunrise in the evening, unlike the daytime.

To this end, although the prior art has used a method of providing a suitable illumination separately, it has been difficult to practically realize shooting of both a person and a vehicle accurately at night using a separate illumination. However, it is now possible to implement a crime prevention system capable of storing moving images and recognizing still images by photographing vehicles and pedestrians in an environment that does not disgust pedestrians by using infrared light.

Hereinafter, embodiments of the integrated security camera system and control method thereof capable of capturing and recognizing both vehicles and persons passing by on the road clearly while photographing a plurality of frames continuously for storage of moving images proposed by the present invention Explain.
Embodiments of the present invention will now be described with reference to the drawings, but this is for a better understanding of the present invention, and the scope of the present invention is not limited thereto.
The integrated security camera system of the present invention is shown in Fig. The security camera system includes a vehicle detection unit 10 for detecting when a vehicle enters, a high-resolution camera 20, a low-resolution camera for capturing moving pictures on the rear side of the vehicle, an infrared lamp 30, a system controller 40 for controlling them, . In one embodiment, the vehicle detection section may include detectors CH1 and CH2 configured as vehicle detection sensors. The system controller 40 includes a video transmission control unit, a vehicle recognition control unit, a vehicle recognition unit, a camera control unit, and a lighting control unit. The camera control unit includes a shutter speed S, an image sensor sensitivity gain G, And the camera resolution (R). The illumination control unit adjusts the brightness (L) of the illumination.
In one embodiment of the present invention, the chanel detection unit 10 uses a loop sensor, a non-contact type laser sensor, or the like for allowing frequency oscillation when the metallic body enters. If there is a problem with the installation of such sensors, a motion detection method, which is a motion detector, can be used at the same time by setting up a virtual loop and determining whether there is an incoming vehicle.
In one embodiment of the present invention, loop sensors (CH1, CH2) for vehicle detection are provided for detecting the passing of a vehicle and the speed of a passing vehicle on the road. In the present invention, a loop sensor, a laser device, or the like can be used to detect the vehicle and detect the speed. The operation principle and the usage state of the loop sensor and the use state are already well known. When the vehicle passes over the loop sensors CH1 and CH2, the loop sensors CH1 and CH2 transmit detection signals for the unpressurized state and the pressed state of the vehicle to the system controller 40. [ The loop sensors CH1 and CH2 and the system controller 40 are connected by wire or wireless according to the method of installation. The system controller 40 detects the vehicle based on the signals inputted from the loop sensors CH1 and CH2 and makes it possible to calculate the speed.
In one embodiment of the present invention, the infrared lamp 30 is designed to have an optical structure that does not interfere with the driver of the driving vehicle and does not interfere with obtaining a clear image by providing an optimal light source to the camera. It is also programmable to trigger the lighting ON / OFF by various methods by PWM (Pulse Width Modulation) control according to the desired resolution

FIG. 2 illustrates a process of acquiring an image through a security camera system corresponding to one step of a camera trigger timing (T) according to an embodiment of the present invention. The number in the circle in FIG. 2 represents the sequence sequence to be processed, and as shown in FIG. 2, the system controller may further include a controller and a control board for operating the camera control unit and the light control unit. In one embodiment of the invention, the control board may be a small server. Hereinafter, a process of acquiring an image by the security camera system will be described.

(1) The control board controls the IO of the camera control unit according to the state of the vehicle detection sensors 1 and 2, generates a trigger signal, and outputs it to the TRIGGER of the camera control unit.

(2) The camera control unit operates the shutter of the camera according to the TRIGGER signal with the set value and outputs the shutter status to the control board.

(3) The control board outputs the illumination brightness information set through the controller to the LEVEL of the illumination control unit and outputs the illumination ON signal to the illumination control unit in synchronization with the shutter state of the camera control unit.

(4) The camera controller sends the acquired image to INTERFACE (GigaE, USB, IEEE1394, etc) of the controller.

(5) The controller recognizes the brightness and sharpness of the image and transmits the camera setting value for optimizing the follow-up to the camera control unit, and the setting value related to the control is transmitted to the communication port of the control board.
In one embodiment of the present invention, the integrated security camera system needs to shoot again when the vehicle passes a certain distance or more with the first loop coil (vehicle detection sensor 1) notifying the entry of the vehicle on the road of the vehicle. (Vehicle detection sensor 2) for detecting the vehicle's license plate, and the license plate photographing image is displayed when the vehicle passes the first loop coil, passes the first and second loop coils at the same time, Only when it has passed.
In an embodiment of the present invention, a loop coil embedded in a road surface may cause a malfunction or disconnection due to traffic or other factors of the vehicle. Therefore, when a failure message is transmitted to the central processing unit It is possible to acquire the license plate image by detecting the motion picture, and when the loop is restored to the normal mode again, the detection mode can be switched again and detected by the loop coil. The results of all these operations can be recorded by the program of the central processing unit.

3 shows a control method of the camera at the time of entering the vehicle. The camera control method before and after the vehicle is detected by the vehicle detection sensor using the loop or virtual loop when the vehicle enters the traveling direction is different from each other.
The photographing conditions before and after the detection of the vehicle in the present invention adjust the shutter speed unlike the conventional technique. So that the lens is controlled by adjusting the shutter speed according to the brightness processed by the program using the information obtained from the vehicle and the object traveling on the road in the so-called brightness estimation, so that two or more shutter speeds are artificially combined Which is different from the method of controlling the shutter speed. The image obtained by the technique of the present invention has a good image sharpness and a wider image than the conventional one, so that it is also possible to view two lanes using one camera. Of course, video storage and vehicle license plate recognition is also possible.

In one embodiment of the present invention, the camera setting is adjusted so that low-illuminance photographing can be performed mainly for object recognition before the detection of the vehicle, and then the camera shutter speed is controlled to be slow. At this time, the brightness of the illumination is made dark so that the life of the illumination can be extended. And, during the detection of the vehicle, the camera setting is adjusted so that a bright and clear image of the moving vehicle can be obtained, and the vehicle license plate tracking algorithm is applied to the acquired image to recognize the license plate. At this time, the camera shutter speed is fast and the lighting is bright. After the detection of the vehicle, the camera and the illumination control mode are set to the same manner as before the detection.
As described above, the security camera system according to the present invention includes a high-resolution camera having a shutter for substantially controlling a time during which an image of an object passes through a lens, for controlling a shutter speed of the camera on the basis of whether or not a vehicle is detected, Such as a camera control unit, which controls the shutter so as to photograph the object continuously with a plurality of frames and the plurality of frames having at least two shutter speeds (shutter speed combinations) while changing the object to at least two shutter speeds And control means.
Here, the shutter means to "substantially" adjust the time that the image of the object passes through the lens. That is, in the case of a mechanical shutter, it means a separate shutter that mechanically adjusts the time that an object image passes through the lens. On the other hand, in the case of an electronic shutter, it means that the signal charge accumulated in the form of an electric charge is adjusted to the image sensor, so that the shutter may not physically exist as a separate shutter. Further, the meaning of the shutter control means means to adjust the shutter speed. That is, in the case of using a mechanical shutter, it means to control a shutter that actually exists mechanically, and may mean, for example, a motor drive. When the electronic shutter is used, the camera control unit controls the shutter of the camera connected to the image sensor processor. The shutter control means generally controls the shutter by the number of shutter operation times per unit time, the shutter speed, and the like.
Although control of the shutter speed can be determined in various ways, the present invention refers to an index table created by statistical data for processing. This table is based on real-time experimental data collected for a long time in the present invention. Because it uses infrared lighting, it is not always possible to activate lighting for video at night. Since there is a problem in ON / OFF lifetime of the infrared LED, the shutter speed according to the brightness of the detected vehicle and object is automatically determined according to the above table. In the place where the system of the present invention is initially installed, the condition information of the surrounding environment is not collected, and a process of learning until standardization is needed. Of course, the above-described shooting conditions are programmed so that the image can be acquired by modifying the shutter speed remotely from the server. It is also possible to change the shooting conditions, change the connection to the camera control unit in the small server, change the shooting conditions in the main server at the remote location, and set the list of shooting conditions as a log file And to be stored in the server 's auxiliary storage device to help track the imaging conditions.
In a further embodiment of the present invention, the shutter speed of the camera is also adjusted depending on the speed of the detection target. Since there is no time to adjust the exposure amount range by adjusting the aperture instantly in real time, it is set to a fast shutter speed according to the index reference table for a fast moving object and to a slower shutter speed for an object moving at a relatively low speed It is possible to acquire a plurality of frames by photographing, and at the same time, it is possible to automatically process license plate information, which is a target recognition target for a still image, by a computer. At this time, the speed calculation is performed by receiving the vehicle detection signals of the loop sensor CH1 and the loop sensor CH2, detecting the time difference [Delta] t thereof, and determining the distance DELTA L between the loop sensors CH1 and CH2, The speed of the vehicle can be calculated.

FIG. 4 shows a vehicle license plate optimization follow algorithm for extracting a license plate from an image obtained through a security camera system. If the license plate area is extracted according to the control of the camera, the camera optimization follow-up method depends on whether the license plate area is recognized or not. If the license plate area is not recognized, the optimal set values of the shutter and gain are set to a value obtained by subtracting the average value from the optimized target value, and the optimized follow-up camera setting value is obtained by multiplying the optimal set value by the current camera set value, Lt; / RTI > On the other hand, if the number area is recognized, the optimization setting value is set to a value obtained by subtracting the recognition area average value from the optimization target value, and the optimization tracking camera setting value is multiplied by the optimization setting value from the current camera setting value, . Finally, the tracking method of the present invention obtains a good quality image by calculating an optimization follow-up set value and applying it.
According to an exemplary embodiment of the present invention, the vehicle tracking algorithm may include parallel processing of the interest core area recognition in the small server to reduce the burden on the computer controller due to the recognition speed of the image obtained by the camera control, And a parallel image processing algorithm for recognizing the unrecognized residual image may be included in the parallel image processing algorithm

5 shows an overall configuration diagram of an embodiment of an integrated security camera system according to the present invention. The following description will be made with reference to Fig.

In the present embodiment, the integrated security camera system includes a camera control unit 80 having a function of substantially capturing an object through a lens. One high-speed camera 20 is mounted, and a high-resolution camera 20 The preprocessed image information and the vehicle image data are transmitted to the main server 70, which is a central processing unit, through the streaming of the small server 90.
The camera control unit 80 includes a 1-chip microcomputer-based control circuit. The small server 90 for storing and transmitting the small video compresses the still image and moving image data output from the camera in the H.264 format and transmits the compressed image data to the central processing unit And is configured to be able to communicate with the main server 70.

The CPU 90 and the operating system (OS) -based small server 90 responsible for the distributed processing and the image transfer are provided with a simple processor for processing a frame, a memory, an auxiliary storage device, an operating system for high-speed processing, Can be used. The miniature server 90 based on the embedded computer can receive images taken through the Internet interface of the camera controller 80 and can preprocess or store them by a program of a pipeline type and transmit them by streaming . In addition, the small server 90 can change and control the shooting conditions such as the shutter speed of the camera controlled by the camera control unit 80 interfaced with the serial communication by program control operated in the operating system.
The mini server 90 also extracts the license plate area of the vehicle from the image of the vehicle, extracts individual characters from the license plate area, and recognizes the extracted characters. The vehicle license plate recognition device captures the entire image of the vehicle when the entering vehicle passes through a loop coil, and only the license plate image is extracted by the computer to automatically recognize the license plate of the entering vehicle. Also, images of the front and full length of the entering vehicle are captured, and images including the license plate are acquired to transfer the images to the storage device. The database includes access times, full image plate partial images, recognized results, and vehicle information. The unit uses infrared light to prevent driver glare and guarantees long life. Also, the maintenance cost can be reduced due to the long service life. Also, a control circuit device is attached to allow the camera to be adjusted according to time and place.
Since the video transmission is performed by the small server 90 with an image of 15 frames or more and the license plate recognition should not be interfered with, the compression process is performed in the H.264 format for high-speed transmission and the RGB video is precisely matched to the transmission format It is important. This process consists of encoding, image conversion, and transmission format matching. In order to handle such a situation, it is only necessary to support the hardware requirements and programming.
The central processing main server 70 can process and store distributed image information and moving images, and can be connected to an intranet managed by a private domain, a police agency, or a local government agency.
According to one embodiment, the security camera system may include a relay driver (not shown), a power controller (not shown), and a camera tree reject (not shown). The relay driving unit receives the control signal from the camera control unit, the power control unit sends and receives signals to the small server to control the power, and the camera tree rejection can transmit the signal to the camera control unit to trigger the high resolution camera.

delete

The camera control unit 80 of the present invention includes an electronic synchronization signal for obtaining images from a camera including a lens, a camera shutter speed control algorithm, a brightness control algorithm for controlling the illumination board, a temperature control of the camera control unit, , An operation diagnostic program, and the like. In addition, the camera control unit constitutes a serial communication mode with a small server.

delete

6 is a block diagram illustrating an overall operating system of the integrated security camera system according to the present invention. The small server 90 and the main server 70 output a user interface screen (UI) to monitor the video information of the vehicle entering through the monitor screen. The main server 70 can monitor the front image of the photographed vehicle, the segmented license plate image, and the recognized license plate information, and the UI manager can display the entire image of the vehicle, the vehicle number, the shooting information, You can see all the necessary information such as the passage time. The main server 70 has a function of storing the captured image and the image processing result in the form of a plurality of images compressed in the H.264 format in order to store the vehicle and the passenger image, Of the video stream.

While the present invention has been described with reference to exemplary embodiments, it is to be understood that the invention is not limited to the disclosed exemplary embodiments.
&Lt; Embodiments of the present invention &
Hereinafter, embodiments of the present invention will be described.
According to an embodiment of the present invention, a security camera system includes a system controller for controlling two vehicle detectors, a high-resolution camera and a low-resolution camera, an infrared lamp, a small server, and a security camera system, , The vehicle detection control, the vehicle recognition, the camera control, and the lamp control function. The system controller includes a lighting control unit. The high resolution camera shoots the license plate. The low resolution camera shoots the rear surface of the vehicle, The high resolution camera is used for LPR (License Plate Recognition) and movie storage. The high resolution camera includes a shutter speed controller. The high resolution camera controls the image sensor sensitivity gain and the camera resolution. The shooting speed of the high resolution camera is 15 frames per second And the two vehicle detectors are loop or virtual loops And the shutter speed controller is set differently according to the camera control method before and after passing through the vehicle detection sensor, and the lens of the high resolution camera is photographed The camera control method before the detection of the vehicle is such that the setting of the high-resolution camera is adjusted so that the low-illuminance photographing can be performed mainly on the object recognition, the shutter speed of the high-resolution camera is slowed In this case, the brightness of the illumination of the infrared lamp is made dark to extend the life of the infrared lamp illumination. During the detection of the vehicle, the setting of the high-resolution camera is adjusted so that a bright and clear image of the moving vehicle can be obtained. The vehicle follow-up algorithm is applied, The camera control method of the camera control method is as follows. The camera control method of the present invention enables to capture images of a plurality of frames, A security camera system includes a camera control unit having a function of substantially capturing an object through a lens, the camera control unit including a camera control unit for acquiring an image from a camera including a lens, The function of the camera is to perform functions including control of the electronic synchronous signal and camera shutter speed, brightness control algorithm for controlling the lighting board, temperature control of the camera control unit, pan tilt control, firmware update and operation diagnosis program. Based on the data, And a further control according to the brightness and rate, when a failure message is sent from the loop, and acquires an image of a vehicle by an image sensing method by the virtual loop.
According to an embodiment of the present invention, the security camera system processes data in the order of a high-resolution camera, a camera controller, and a small server, a camera controller transmits a control signal to a relay driver, The small server transmits and receives signals to and from the integrated security management server via the Internet network. The camera control unit transmits a signal to the lighting control unit for controlling the illumination, The system includes a camera tree rejection, and the camera tree rejection sends a signal to the camera control to trigger a high resolution camera.
According to an embodiment of the present invention, in the security camera system, two or more vehicle detection sensors are installed, the system controller includes a control board and a first controller, the signal of the vehicle detection sensor is input to the control board, The board communicates with the first controller, the control board gives a signal to the lighting control unit to turn on / off the lighting, the control board gives the trigger signal and the input signal to the camera control unit, the camera control unit gives the strobe signal to the control board, The camera control unit performs bidirectional communication with the first controller and the USB port.
According to an embodiment of the present invention, in the security camera system, the vehicle tracking algorithm includes: detecting the vehicle based on the vehicle detection sensor; Extracting a license plate expected area according to the detected vehicle; Recognizing a number area; And the number area is not recognized, the first set value of the shutter and gain of the high resolution camera is set to a value obtained by subtracting the expected area average value from the target value, and the follow camera setting value of the high resolution camera is set to the value 1 set value, and the small server is connected to the main server, which is a central processing system, through the communication network.
According to an embodiment of the present invention, in the security camera system, the vehicle tracking algorithm includes: detecting the vehicle based on the vehicle detection sensor; Extracting a license plate expected area according to the detected vehicle; Recognizing a number area; And the number area is recognized, the first set value of the high resolution camera is set to a value obtained by subtracting the average value of the recognition area from the target value, and the follow camera setting value of the high resolution camera is obtained by multiplying the current set value of the high resolution camera by the first set value And the system controller includes a video transmission control unit, a vehicle recognition control unit, a vehicle recognition unit, a camera control unit, and a lighting control unit, and the camera control unit adjusts the shutter speed, the image sensor sensitivity gain and the camera resolution, The control unit continuously captures the object in a plurality of frames by adjusting the brightness of the illumination, stores the captured image, and the small server compresses the moving image data in the H.264 format and performs transmission.
According to an embodiment of the present invention, in a security camera system, a vehicle tracking algorithm is a method in which only a key region recognition is parallel-processed in a small server to reduce the burden of a computer controller due to a recognition speed, And a parallel image processing algorithm for transmitting the region of interest information and the original image obtained from the camera together to the main server installed in the central control center and for recognizing the unrecognized residual image.

10: vehicle detection unit
20: High resolution camera
30: Infrared lamp
40: System controller
60: Broadband network
70: Integrated Security Management Main Server
80: camera control unit
90: Small server

Claims

In a security camera system,
Two vehicle detectors;
High resolution camera and low resolution camera;
Infrared lamp;
Small servers; And
And a system controller for controlling the security camera system,
Wherein the system controller has a video image transmission control, a vehicle detection control, a vehicle recognition, a camera control, a lamp control function, and the system controller includes a lighting control unit,
The high-resolution camera photographs a license plate, the low-resolution camera photographs a rear surface of the vehicle, stores moving images,
Wherein the high resolution camera is used for LPR (License Plate Recognition) and motion picture storage, and the high resolution camera includes a shutter speed controller; Wherein the high resolution camera adjusts an image sensor sensitivity gain and a camera resolution, the shooting speed of the high resolution camera is at least 15 frames per second,
Wherein the two vehicle detectors comprise a vehicle detection sensor using a loop or virtual loop, the image detection method being used when the virtual loop is used,
The shutter speed regulator is set differently according to the camera control method before and after passing through the vehicle detection sensor and is determined on the basis of the number plate of the object or vehicle in which the object is photographed in the shooting range photographed by the lens of the high-
The camera control method before the vehicle detection adjusts the setting of the high-resolution camera so that low-illuminance photographing can be performed mainly with object recognition, and then the shutter speed of the high-resolution camera is controlled to be slow. At this time, Resolution camera is adjusted so that a bright and clear image of the moving vehicle can be obtained during the detection of the vehicle while the vehicle is being detected, and the vehicle tracking algorithm is applied using the acquired image. At this time, the shutter of the high- The speed is set to a high speed, the illumination is brightly adjusted, and after the detection of the vehicle,
The camera control method includes acquiring a license plate from a still image by enabling image acquisition of a plurality of frames, and simultaneously transmitting moving images of a continuous frame,
The security camera system includes a camera controller having a function of photographing an object through a lens,
The camera control unit includes an electronic synchronization signal for obtaining an image from a camera including a lens, a camera shutter speed control, a brightness control algorithm for controlling the illumination board, a temperature control of the camera control unit, a pan tilt control, a firmware update, To perform the function of,
The shutter speed of the high resolution camera is further controlled in accordance with the brightness and speed of the detection subject based on the statistical data,
Wherein when the failure message is transmitted from the loop, the image of the vehicle is acquired by the image sensing method by the virtual loop.

delete

The method according to claim 1,
The security camera system includes the high-resolution camera; The camera controller; The data is processed in the order of the small server, and the camera control unit transmits a control signal to the relay driving unit; Wherein the small server transmits and receives signals to and from a power control unit for controlling power supply;
The small server is configured to send and receive signals to the integrated security management server through the Internet network; The camera control unit transmits a signal to the illumination control unit for controlling the illumination, the security camera system includes a camera tree rejection, and the camera tree rejection transmits a signal to the camera control unit to trigger the high resolution camera Wherein the security camera system comprises:

The method according to claim 1,
Wherein the system controller includes a control board and a first controller, a signal of the vehicle detection sensor is input to the control board, and the control board communicates with the first controller , The control board gives a signal to the illumination control unit to turn on / off the illumination, the control board gives a trigger signal and an input signal to the camera control unit, the camera control unit gives a strobe signal to the control board, Wherein the camera controller is configured to bidirectionally communicate with the first controller through the USB port.

2. The method according to claim 1,
Detecting the vehicle based on the vehicle detection sensor;
Extracting a license plate expected area according to the detected vehicle;
Recognizing a number area; And
The first set value of the shutter and gain of the high resolution camera is set to a value obtained by subtracting the expected area average value from the target value and the follow camera setting value of the high resolution camera is set to the current setting of the high resolution camera And multiplying the first set value by the first set value,
Wherein the small server is connected to a main server which is a central processing system through a communication network.

9. The method of claim 8, wherein the vehicle tracking algorithm
Detecting the vehicle based on the vehicle detection sensor;
Extracting a license plate expected area according to the detected vehicle;
Recognizing a number area; And
Wherein the first setting value of the high resolution camera is set to a value obtained by subtracting the average value of the recognition area from the target value and the following camera setting value of the high resolution camera is set to the first setting value of the high resolution camera, Determining a value multiplied by the set value,
The system controller includes a video transmission control unit, a vehicle recognition control unit, a vehicle recognition unit, the camera control unit, and the illumination control unit. The camera control unit adjusts a shutter speed, an image sensor sensitivity gain and a camera resolution, Wherein the small server sequentially captures the object in a plurality of frames by adjusting a brightness level of the object and stores the captured image, and the small server compresses the moving image data into H.264 format and performs transmission.

delete

12. The method according to claim 11, wherein the vehicle tracking algorithm includes parallel processing of the interest center area recognition in the small server to reduce the burden on the computer controller due to the recognition speed of the image obtained by the camera control, And a parallel image processing algorithm for transmitting the obtained original image together with the main server installed in the central control center and recognizing the unrecognized residual image.