KR101322162B1 - Method for detection using detection system of vehicles - Google Patents

Abstract

Vehicle detection method using a vehicle detection system according to the present invention is a vehicle that maximizes the efficiency of the detection of the vehicle and the vehicle number, the vehicle running as well as the lane change of the vehicle with the intention to avoid detection or reverse driving beyond the center lane It is also possible to take pictures of traffic, so it is possible to effectively crack down on traffic violations.

Description

Method for Detection using Detection System of Vehicles

The present invention relates to a vehicle detection method using a vehicle detection system, and more particularly, to improve the performance of detecting, passing through the image pickup and vehicle number recognition of vehicles passing through the monitoring area of the main road by using an image photographing camera. The present invention relates to a method of detecting a vehicle.

In general, a checkup or search conducted for the arrest of a vehicle used for crime or a criminal on board a crime vehicle usually checks the vehicle number visually after stopping a vehicle passing by a police officer and uses a radio to determine whether the vehicle is a crime vehicle. The method of identification was used. However, such a method is inefficient in terms of manpower and time for inspecting or searching a vehicle one by one due to an increase in the number of vehicles and the traffic volume, and a problem of causing traffic jams due to the inspection or search is emerging.

Recently, crime prevention vehicles such as wanted vehicles, crime vehicles, stolen vehicles, arrears, or traffic violations have been installed by installing crime prevention vehicle identification systems on major roads. Coping with

Conventional anti-theft vehicle number recognition system is applying a variety of devices or methods for detecting the vehicle. For example, roof coils, vehicle detectors, Auto IRIS lenses, digital CCD cameras of over one million pixels, strobe lighting devices, image frame grab boards, vehicle identification key controllers, and the like.

As described above, according to the conventional vehicle number recognition system, the roof coil detection method embeds a roof coil on the road to detect a vehicle traveling on the road, and receives a signal of the vehicle detector connected to the roof coil to receive the signal of the vehicle number recognition main controller. Take a vehicle image to perform number recognition. The digital CCD camera is installed to be linked to the infrared strobe to take a picture at the same time as driving the infrared strobe.

However, the buried or contact roof coil (detection area; 1,800mm × 1,800mm) for detecting the vehicle has to be dug into the road because the roof coil must be installed on the road surface, so there is a risk and inconvenience in the installation work. There is a problem in that the life of the roof coil is limited, and the cost of the roof coil is increased. In addition, the loop coil needs to reset the sensor sensitivity at about 6 month intervals, and if a failure or loss of the loop coil occurs, the replacement or repair is not easy, thus increasing the cost for maintenance.

Therefore, to compensate for the problem of the roof coil, a laser detection method for detecting a vehicle using a non-embedded and non-contact laser sensor has been introduced. This method has the advantage that the detection point can be operated variably during the day, can be used for a long time, and there is no possibility of disconnection by road construction. However, the detection rate is remarkably lowered due to scattering of the laser light due to yellow dust, rainfall, or snowfall, the position of the vehicle detection is changed by the influence of vibration such as wind, cosine error exists, and it is vulnerable to fog. Continuous reset is required, and the equipment is expensive, especially the detection area of the laser point has a narrow problem of about 30mm or less.

Recently, an image detection method has been applied to compensate for the problems of the laser sensor. This method is provided with a surveillance camera for photographing a surveillance area, a number recognition camera for photographing a vehicle detected from the surveillance image of the surveillance camera, and an infrared LED lighting device which is driven to emit light in synchronization with the surveillance camera and the number recognition camera. As such, the video sensing device using the camera reads the surveillance video of the surveillance camera, checks the appearance of the vehicle, controls the driving of the number recognition camera and the infrared LED light so that the vehicle can be photographed by the number recognition camera while tracking the vehicle in the surveillance image. And a vehicle number recognition main controller that recognizes the vehicle number in the captured image frame, and the vehicle number recognition main controller divides the surveillance image into a reading area and a non-reading area and monitors only the reading area to read whether the vehicle is present. .

Such an image sensing device has a detection area of 3,700 mm x 7,000 mm x 6000 mm in width x length x height in a three-dimensional detection method. This method does not require constant resetting of the detection area, is easy to maintain and semipermanent. In addition, the possibility of disconnection due to road construction is excluded, and there is an advantage that little maintenance costs are incurred. However, there is a problem in that a vehicle missed during the recognition of the vehicle number and the load of the system for continuously recognizing the license plate number by taking a still image from the video stream and recognizing the license plate number.

Among the methods of detecting the vehicle, the loop coil detection method is the highest, followed by the laser detection method, the image detection method through image analysis, and the detection method for recognizing the vehicle number. .

Based on the detection area of the vehicle, the one-dimensional laser detection method (detection radius is 30 mm or less) is the narrowest, and the two-dimensional loop coil detection method (detection area is 1,800 mm × 1,800 mm), and three-dimensional It is the order of detecting vehicle number and image detection method through image analysis (interest detection area is 3,700mm × 7,000mm × 6,000mm).

And based on the difficulty and cost of maintenance, the detection method for recognizing the vehicle number and the image detection method through image analysis are the lowest, followed by the laser detection method and the loop coil detection method.

As such, it is the accuracy of the vehicle detection that increases the reliability of the security CCTV system for vehicle number recognition. The accuracy of vehicle detection is the order of roof coil detection method, laser detection method, image detection method through image analysis, and vehicle number recognition detection method. However, when the criminal vehicle intends to evade the camera shooting, the vehicle detection area is very important. If the accuracy of vehicle detection is high, but the detection area is narrow and can easily avoid shooting, the existing vehicle identification security CCTV system had a meaningless installation and application itself.

The present invention is to solve the above problems, to maximize the advantages of the corresponding detection method through the image analysis method using a single CCD camera to facilitate the acquisition of the vehicle and vehicle number and acquisition of the captured image Purpose.

Furthermore, in order to avoid the detection of the vehicle by the driver of the main road, the driver can detect and photograph the vehicle even when the lane is changed intentionally or if the driver reverses the center line without permission. Another purpose.

In order to achieve the above object, the present invention provides a vehicle detecting apparatus including an image photographing camera for detecting a vehicle and capturing an image; A receiver for controlling input and output of data generated by the vehicle detecting apparatus; A local server for controlling the vehicle detection signal and the image data inputted from the receiver to recognize, store, and transmit and receive the vehicle and the vehicle number; A vehicle security operation server for transmitting and receiving vehicle traffic information, vehicle information, and environment setting information through the control local server and a communication network; A vehicle number recognition server for recognizing a vehicle number among vehicle information received through a communication network from the control local server; A field image display server for visually displaying an image of a field received through a communication network from the control local server; A DB and a data processing server for storing, processing, and calculating various information and data received through a communication network in the local server for control; Police station DB for providing theft or wanted vehicle information stored in the police station through the communication network; In the vehicle detection method using a vehicle detection system comprising a police station terminal for transmitting and receiving data to be installed at the police station located in the field and a short distance through the control local server and a communication network (a) the image recording camera control unit of the control local server Setting a region of interest (ROI) corresponding to a logical vehicle detection and tracking section among image areas continuously photographed by a CCD camera of the photographing camera; (b) setting a logical photographing first reference region and a second reference region to accurately photograph a plurality of evidence photographs of interest regardless of speed with respect to a vehicle traveling in the region of interest; (c) After setting up the ROI and the reference region, image analysis is performed only within the ROI in each image frame photographed continuously by a CCD camera to detect a vehicle region, and a minimum neighboring rectangle including the detected vehicle region (MBR). Outputting a Minimum Bounding Rectangle); (d) Calculating the displacement of the center point coordinates of the smallest rectangles in the ROI of the current frame of the image frame and the center point coordinates of the smallest rectangles in the ROI of the image frame up to the previous point in the vehicle region detection process. Making; (e) determining whether a coordinate of the center point of the least adjacent rectangle is within a region of interest; (f) photographing a vehicle in a plurality of reference areas when the center point coordinates deviate from the region of interest and make a left side deviation or lane change that invades a center line or a right side violation that invade a shoulder; (g) storing the photographs taken with the photographic evidence and related data in a DB and a storage unit; (h) characterized in that it provides a vehicle detection method using a vehicle detection system comprising the step of transmitting the photographs taken with the photographic evidence and the related data through a communication network to a remote center system.

According to the present invention, by detecting the vehicle only in the region of interest in the image region photographed by the camera, unnecessary calculation amount of unnecessary image processing can be minimized, so that the vehicle can be detected and tracked in real time.

In addition, by detecting the vehicle by using an image photographing camera of the vehicle detection device to enable accurate vehicle detection, which is the most important in crime prevention vehicle number recognition system.

In addition, since the image detection and vehicle tracking technology having a large detection area can photograph not only a vehicle traveling in a steady direction but also a vehicle traveling in a reverse lane invading the central lane, there is an advantage that it can effectively control a vehicle that violates a traffic law.

1 is a block diagram showing the configuration of a vehicle detection system according to the present invention.
2 is a view schematically showing the installation and operation of the vehicle detection system of the present invention.
3A to 3C are diagrams schematically illustrating detecting a vehicle normally driving in a lane in the vehicle detection system according to the present invention.
4A to 4C are diagrams schematically illustrating detecting a vehicle traveling in a lane in a vehicle detection system according to the present invention.
5A to 5G are diagrams schematically showing a method of detecting a vehicle in the present invention.
6A and 6B are flowcharts illustrating a vehicle detection method using the vehicle detection system of the present invention.

Hereinafter, a vehicle detection method using the vehicle detection system of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram showing the configuration of a vehicle detection system according to the present invention.

First, the vehicle detection system of the present invention sets a region of interest (ROI) from an image acquired from a camera, sets a minimum adjacent rectangle (MBR) of the vehicle in the region of interest, and calculates a center point coordinate (CP) at the minimum adjacent rectangle. And to determine whether or not the center point coordinates deviate from the region of interest so as to search, discriminate, and track the missed vehicle or criminal offense or the violation subject to the captured vehicle information.

The vehicle detecting apparatus includes an image photographing camera 200 which detects and photographs a vehicle. Furthermore, the image photographing camera 200 detects a vehicle and photographs an image. The image photographing camera 200 includes a vehicle detection and number recognition CCD camera 230 for photographing a vehicle's detection and a vehicle number. The vehicle detection and number recognition CCD camera 230 is provided as a single unit but may be provided as a plurality. Further, the CCD camera 230 for vehicle detection and number recognition is installed in the cradle 5 by car, and installed in both the forward driving direction and the reverse driving direction, respectively, and the CCD camera 230 installed in the forward driving direction or the reverse driving direction of the road. Is installed to photograph the rear of the vehicle running in the relative lane.

In addition, the image photographing camera 200 includes an infrared LED lighting device 210 having a plurality of infrared LEDs that are driven to emit light by an optical sensor CdS in order to obtain an image frame of the CCD camera 230 according to the ambient light. Included. The infrared LED lighting device 210 controls infrared (IR) light emitting diode (LED) illumination by a light sensor (CdS). In other words, the operation of the infrared LED lighting device 210 is controlled according to the ambient illumination. In addition, the image recording camera 200 includes a motorized lens 220 for adjusting the focus and distance for the image of the object to be photographed, the rotation and tilt of the image recording camera 200, that is, pan tilt (Pan / Tilt) In order to adjust the image taking camera 200, the camera direction switching unit 300 is installed.

In FIG. 2, the vehicle detecting apparatus is installed in the cradle 5 between the lanes 2 of the road 1 or across the one-way or reciprocating lanes, respectively. The detection signal detected by the image capturing camera 200 is transmitted to the control local server 500 via the receiver 400.

The receiver 400 controls input / output of data generated by the image photographing camera 200 of the vehicle detecting apparatus. The receiver 400 performs the processing of the input / output data, the input / output or the control signal, or the transmission / reception, and is usually installed in a control box or the like.

The control local server 500 analyzes the vehicle detection signal and the image data input from the receiver 400 to recognize, store and transmit the vehicle and the vehicle number. The local server 500 for control includes an image photographing camera control unit 510 which analyzes the illuminance of the site with the image of the vehicle and various information data input from the receiver 400 and controls the operation of the vehicle detecting apparatus. In addition, the image analyzer 520 of the local server 500 for controlling is configured to set a region of interest from the image information received from the image capturing camera controller 510, classify a photographing line, detect a vehicle, track a vehicle, and capture an image. The process of recognizing the vehicle number is performed. The image analyzer 520 includes a program for detecting and determining whether the vehicle leaves the left side or the right side of the ROI. The image analyzer 520 includes a module for analyzing image information.

In addition, the control local server 500 includes a vehicle number recognition unit 530 that interprets the vehicle number recognized by the image analyzer 520, and information input and output to the control local server 500 and driving of each component. DB and a storage unit 540 for storing the generated information. In addition, a communication unit 550 for performing communication for transmitting and receiving information data with the field system or the center system is included. Multiple field systems are installed remotely from the center system. Therefore, the sharing and transmission of information generated in the field system is required, so data will be transmitted and received through wired, wireless or wired and wireless communication networks.

Moreover, the on-site system transmits cumulative information on the on-site vehicle traffic, requesting a list of crime vehicles, lane departure information of the vehicle, reporting the appearance of a crime vehicle and the detected vehicle number, and transmitting images of crime vehicles and passengers to the center system. In addition, the center system transmits and controls the environment setting work for providing the list of criminal vehicles, updating the occasional or periodic DB, etc. to the field system.

The center system includes a plurality of management and control servers for the control of the field system and the transmission and reception of data, and in particular, it is linked with servers such as public institutions and government offices for data sharing.

The vehicle security operation server 600 of the center system transmits and receives traffic information, vehicle information, and environment setting information of the vehicle through a local network 500 for control and a communication network. And the vehicle number recognition server 700 is to recognize the vehicle number from the vehicle information received through the communication network from the local server for control 500, the police agency DB 800 provides theft or arrangement vehicle information stored in the police station through the communication network The on-site image display server 900 visually displays the image of the field received through the communication network from the local server 500 for control, and displays the image of the field and map information through a plurality of monitors. . The DB and data processing server 1000 is a server for storing, processing, and calculating various information and data received through a communication network from the local server 500 for control in a DB. In addition, the police station terminal 1100 is installed in a network at a police station located on the site and a short distance through a local network for control 500 and a communication network to transmit and receive data.

As described above, the vehicle detection system of the present invention enables to quickly and accurately track and acquire vehicle information traveling on a road using a vehicle detection device having various types and characteristics. Next, FIGS. 2 to 5 and 6 will be described. It will be described with reference to the flowchart.

First, the image capturing camera controller 510 of the local server 500 for controlling is an ROI corresponding to a logical vehicle detection and tracking section among the image regions continuously photographed by the CCD camera 230 of the image capturing camera 200. , Region of Interest) is set (S11).

In addition, a logical photographing first reference region T1 and a second reference region T2 are set to accurately photograph a plurality of evidence photographs of interest regardless of speed with respect to a vehicle driving in the ROI. (S12).

After setting the ROI and the reference regions T1 and T2, the image photographing camera 200 of the vehicle detecting apparatus detects the vehicle (S13). In this case, the control local server 500 controls the driving of the image photographing camera 200 to detect the vehicle.

When the detection of the vehicle is made, the control local server 500 continuously photographs the vehicle with the CCD camera 230 (S14), and performs image analysis only within the region of interest in each image frame to detect the vehicle region. And outputs a minimum bounding rectangle (MBR) including the detected vehicle area (S16). In the vehicle area detection process, the displacements of the center point coordinates of the smallest neighboring rectangle in the ROI of the current view image frame and the center point coordinates (CP, Center Point) of the smallest neighboring rectangle in the ROI of the image frame up to the previous point of time are calculated ( S17).

In operation S18, it is determined whether the center point coordinate CP of the least adjacent rectangle MBR is in the ROI. The vehicle is photographed in a plurality of reference areas T1 and T2 when the center point coordinate CP moves away from the ROI to the left side deviation or lane change invading the center line 6 or the right side deviation to invade the shoulder. (S19). That is, in the process of tracking the vehicle, the smallest rectangle of the vehicle takes a logical photograph in the ROI of the image frame. The first reference area T1, the first reference area T1, the second reference area T2, and the second reference When passing through the area T2, the photograph of evidence of the license plate, the driver, the loading box, etc. of the driving vehicle is taken regardless of the driving speed of the vehicle.

In FIGS. 3A to 3C, a vehicle that normally travels without departing from the lane 2 and the center line 6 passes through the reference areas T1 and T2, where general shooting is performed by the image photographing camera 201. However, in FIGS. 4A to 4C, the vehicle that departs from the lane 2 or the center line 6 and reverse-drives the relative lane sequentially passes the rear of the vehicle in the image photographing camera 202 while passing through the reference areas T3 and T4. Shoot. Therefore, when the vehicle reverses driving without the normal lane, the image photographing cameras 201 and 202 detecting the vehicle in the other lane may acquire the vehicle information by photographing the illegally driving vehicle.

5A through 5C illustrate a process in which the actual image capturing cameras 201 and 202 detect a vehicle. When the image photographing cameras 201 and 202 are described based on a corresponding lane, that is, a left lane, the vehicle 3 approaches the ROI after the ROI is set on the road. Set the minimum neighboring square of (MBR) and calculate and set the center point coordinate (CP) of the minimum adjacent square (MBR). Accordingly, the image capturing camera control unit 510 and the image analyzing unit 520 of the control local server 500 receive the smallest rectangular (MBR) and center point coordinates (CP) of the vehicle 3 received from the image capturing cameras 201 and 202. ) Detects and determines whether the vehicle is normally traveling in the ROI. Therefore, the vehicle detects and determines the displacement of the vehicle to the left side or the right side of the ROI set based on the center point coordinate CP of the vehicle.

Furthermore, FIG. 5D shows whether the vehicle driving on the left side of the lane is driving backwards beyond the center line 6 or in FIG. 5E, whether the vehicle is already backing on the opposite side of the center. It can be judged by the detection of the displacement. FIG. 5F detects and determines that the center point coordinates CP of the smallest rectangle MBR set in the ROI of each vehicle are normally driven without deviation when the plurality of vehicles travel along the roadway. 5G is for detecting and determining a vehicle that is going to cross the center line or a vehicle that has already reversed the center line in a situation where a plurality of vehicles are traveling along the lane.

On the other hand, the photographs taken as evidence photographs and related data are stored in the DB and storage unit 540 (S20). Then, the photographs taken as evidence photographs and related materials are transmitted to the remote center system through the communication unit 550 (S21).

And when receiving the data from the vehicle security server of the center system through the communication network (S22), the vehicle security operation server 600 to recognize the license plate from the vehicle number recognition server 700 from the received license plate photo (S23) Compare the criminal vehicle license plate strings stored in the police station DB 800 and the string of the recognized license plate (S24), and checks whether the crime vehicle has passed, and automatically transmits and reports the identified crime vehicle to the police station terminal (1100) ( S25), based on the character string and the passing time of the license plate, automatically search for the movement route for the vehicle of interest and display it on the monitor through the field image display server 900 (S26), DB and data processing server (1000) After the various crime prevention data and data to be stored and recorded (S27).

As described above, the vehicle detection system of the present invention detects a vehicle changing lanes and shoulders by avoiding photographing in a lane because the detection area is wide by 3,700 mm × 7,000 mm × 6,000 mm in width, length, and height, respectively, in a three-dimensional detection method. can do. Moreover, maintenance is easy and semi-permanent. There is no possibility of disconnection due to road construction and there is no additional cost. Image detection is possible even in the reaction of sensitive natural environment, which is a disadvantage of laser method. In addition, there is an advantage of minimizing the amount of problems that can be missed while recognizing the vehicle number and the load of the system for continuously performing the vehicle number recognition of the vehicle number recognition camera through the detection of the vehicle number through image detection. .

While the invention has been shown and described in connection with specific embodiments thereof, it is well known in the art that various modifications and changes can be made without departing from the spirit and scope of the invention as indicated by the claims. Anyone who has a can easily know.

200: Video recording camera 210: LED lighting device
220: motorized lens 230: CCD camera
300: camera direction switching unit 400: receiver
500: local server

Claims (5)

A vehicle detecting apparatus including an image photographing camera for detecting a vehicle and capturing an image; A receiver for controlling input and output of data generated by the vehicle detecting apparatus; A local server for controlling the vehicle detection signal and the image data inputted from the receiver to recognize, store, and transmit and receive the vehicle and the vehicle number; A vehicle security operation server for transmitting and receiving vehicle traffic information, vehicle information, and environment setting information through the control local server and a communication network; A vehicle number recognition server for recognizing a vehicle number among vehicle information received through a communication network from the control local server; A field image display server for visually displaying an image of a field received through a communication network from the control local server; A DB and a data processing server for storing, processing, and calculating various information and data received through a communication network in the local server for control; Police station DB for providing theft or wanted vehicle information stored in the police station through the communication network; In the vehicle detection method using a vehicle detection system comprising a police station terminal which is installed at the police station located in the field and a short distance through the control local server and a communication network, transmitting and receiving data,
(a) setting a region of interest (ROI) corresponding to a logical vehicle detection and tracking section among image regions continuously photographed by a CCD camera of the image photographing camera;
(b) setting a logical photographing first reference region and a second reference region to accurately photograph a plurality of evidence photographs of interest regardless of speed with respect to a vehicle traveling in the region of interest;
(c) After setting up the ROI and the reference region, image analysis is performed only within the ROI in each image frame photographed continuously by a CCD camera to detect a vehicle region, and a minimum neighboring rectangle including the detected vehicle region (MBR). Outputting a Minimum Bounding Rectangle);
(d) Calculating the displacement of the center point coordinates of the smallest rectangles in the ROI of the current frame of the image frame and the center point coordinates of the smallest rectangles in the ROI of the image frame up to the previous point in the vehicle region detection process. Making;
(e) determining whether a coordinate of the center point of the least adjacent rectangle is within a region of interest;
(f) photographing a vehicle in a plurality of reference areas when the center point coordinates deviate from the region of interest and make a left side deviation or lane change that invades a center line or a right side violation that invade a shoulder;
(g) storing the photographs taken with the photographic evidence and related data in a DB and a storage unit;
(h) a vehicle detection method using a vehicle detection system comprising the step of transmitting the photographs taken with the photographic evidence and the related data to a remote center system via a communication network. The method of claim 1, wherein the video recording camera,
Vehicle detection and number recognition CCD camera for photographing the vehicle number and vehicle number, and a plurality of infrared LED driven by the light sensor (CdS) in order to obtain the image frame of the CCD camera according to the ambient illumination Infrared LED lighting device, including a motorized lens for adjusting the focus and distance to the image of the subject, using a vehicle detection system equipped with a camera direction switching unit for adjusting the Pan / Tilt of the video camera Vehicle detection method. According to claim 2, The vehicle detection and number recognition CCD camera,
Vehicle detection method using a vehicle detection system that is installed in the cradle by lanes, and installed in both forward driving direction and reverse driving direction, respectively, and the CCD camera installed in the forward driving direction or the reverse driving direction is used to photograph the rear surface of the vehicle traveling in the opposite lane. . The method of claim 1, wherein the control local server,
Image capture camera control unit for analyzing the illuminance of the site by the image of the vehicle and various information data input from the receiver and control the operation of the vehicle detection device, and setting and shooting the region of interest from the image information received from the image capture camera control unit An image analysis unit for classifying lines, detecting a vehicle, tracking a vehicle, capturing an image, and recognizing a vehicle number, a vehicle number recognition unit for interpreting a vehicle number recognized by the image analyzer, and the control local server. Vehicle including a DB and a storage unit for storing the information input and output to and from the information generated by the driving of each component, and a communication unit for performing communication for the transmission and reception of information data with the field system or the center system Detection method. The method of claim 1,
When receiving data from the vehicle security server of the center system through the communication network, the vehicle security server to recognize the license plate from the license plate recognition server from the received license plate pictures, the string of the recognized license plate and the criminal vehicle license plate string stored in the National Police Agency DB Compare and verify the passing of the crime vehicle, and automatically send and report the identified crime vehicle to the police station terminal, and automatically search the moving route for the vehicle of interest based on the string of the license plate and the passing time. The vehicle detection method using a vehicle detection system further comprising the step of displaying the map on the monitor, and storing and recording various crime prevention data and data through the DB and data processing server.

Images