KR100578041B1 - Network camera apparatus having a traffic information collection and addition function and the traffic information process method - Google Patents

Abstract

The present invention provides installation and maintenance by providing a video camera for collecting vehicle traffic image information on a road, an image detector for recognizing and analyzing the collected image information, and a function module for providing various additional functions. The present invention relates to a network camera device having a traffic information collection and additional function, and to a traffic information processing method thereof, which can easily and variously process traffic information.

To this end, the present invention, the camera module unit having a video camera to capture the image of the road, a digital signal processing unit for collecting the traffic information by analyzing the image information captured from the camera module unit, and the traffic information from the digital signal processing unit The network processing unit receives real-time video information from the camera module unit and compresses it and transmits it to a vehicle detection system server (VDS Server) of the traffic control center by a predetermined communication method. A function module unit for transmitting traffic information for transmitting traffic information, providing a text service, recognizing a climate change, and a wanted vehicle is integrally formed;

Also, a process of setting a detection area for a traffic scene screen, a process of extracting a vehicle target by a hybrid extraction method by line / area segments from the set detection area, and a process of tracking and recognizing the extracted vehicle target According to the tracking and recognition of the vehicle target, there is provided a traffic information processing method comprising a process in which the extracted traffic information and image information transmitted from the camera module unit are transmitted to and analyzed by the traffic control frame.

Traffic information, CCTV, camera, video detector, network, traffic control center.

Description

Network camera apparatus having a traffic information collection and addition function and the traffic information process method}

1 is a block diagram of a network camera apparatus having traffic information collection and additional functions according to a preferred embodiment of the present invention;

FIG. 2 is a block diagram illustrating in detail the digital signal processor and the network processor of FIG. 1;

3 is a detailed block diagram of a vehicle identification unit of FIG. 2;

4A and 4B are schematic diagrams showing a state in which the actual road image information is transformed into a detection area according to the present invention;

5 is a view showing an embodiment in which a network camera device of the present invention is installed on a road;

6 is an overall flowchart illustrating a method of processing traffic information collected by a network camera apparatus for collecting traffic information according to an exemplary embodiment of the present invention;

7 is a flowchart for setting a detection area by converting a real world information coordinate system to an image coordinate system according to the present invention;

8 is a flow chart illustrating extraction of a vehicle target in accordance with the present invention;

FIG. 9 is a flowchart illustrating tracking and recognizing a vehicle target extracted by FIG. 8.

* Description of the symbols for the main parts of the drawings *

100: network camera, 102: camera module unit,

110: digital signal processing unit, 112: digital signal control unit,

120: network processing unit, 122: network control unit,

130: communication module unit, 132: wireless LAN module,

134: wired LAN module, 136: mobile communication module,

140: power supply unit, 150: accidental situation processing unit,

160: traffic information transmission unit, 170: character service execution unit,

180: climate sensor interface unit, 190: arrangement vehicle recognition unit,

200: traffic control center, 300: mobile communication network,

400: police terminal.

The present invention relates to a network camera device having traffic information collection and additional functions, and a traffic information processing method thereof, and more particularly, to collecting various traffic information on a road in real time and using the traffic information through various wired / wireless communication networks. The surveillance camera, image detector, and function module unit that can provide various additional functions are provided together with miniaturization and integration, so that traffic information collection and additional functions can be easily installed at predetermined places and the system construction cost can be reduced. It relates to a network camera device having a traffic information processing method thereof.

As is well known, the demand for vehicles and transportation is increasing due to the expansion of social and economic scales due to industrialization, but the actual road supply rate is not up to the demand and supply of increasing vehicles. Due to traffic congestion costs and casualties, individuals and countries bear enormous social costs.

Such traffic congestion increases the logistics cost, weakening external competitiveness, and causing a great burden on the national economy because it is caused by environmental problems such as pollution due to air pollution and social problems such as traffic accidents.

In order to solve this problem, the necessity of traffic information for efficient operation and management of the roads in accordance with the increase rate of traffic demand is increasing. Therefore, the past traffic through facility investment business on the supply side and demand management policy on the demand side Breaking away from problem-solving solutions, the Intelligent Transport System (ITS) for new traffic management and management policies is in the spotlight.

The intelligent transportation system (ITS) is a next-generation traffic system that combines the existing transportation system with intelligent technologies such as electronics, information, communication, and control, and is provided to drivers, pedestrians, and traffic facility operators and managers to control traffic, operation, and management. Provide exchange information services for efficient measures.

In other words, it analyzes traffic information such as vehicle characteristics and speed on the road, implements road traffic management and optimal signal system, and provides various traffic information to drivers in real time to ensure safe and smooth road driving. By identifying the operating conditions and supporting the optimal operation according to the road conditions, and providing information such as various traffic conditions or traffic accidents to increase the ability of road communication.

On the other hand, in order to provide the above-described traffic information service, road traffic information such as road traffic volume, vehicle driving speed, road occupancy rate, queue length, traffic accident, etc. must be accurately collected.

In order to collect the traffic information, a suitable method is applied according to the conditions of each road. Typically, a method of collecting on-site traffic conditions using a surveillance camera (CCTV) and a change in inductance of a loop coil installed on a road A loop detection method for collecting road information, a method using an image detector for analyzing vehicle image information collected by a camera, and a method for collecting traffic information through a traffic information collecting terminal mounted on a vehicle or a beacon installed on a road This is used.

This will be described in more detail.

First, according to the surveillance camera method, a video of the road traffic scene is photographed by the surveillance camera, and the captured video information is transmitted to the traffic control center via an optical communication cable.

The traffic control center is composed of an image detector, a vehicle detection system server (VDS Server) and a display device, and the video information is collected from the traffic control center as traffic volume, speed, occupancy rate and other traffic information.

Secondly, the traffic information collection method using the loop detector determines whether a vehicle is present due to the change in inductance of the loop coil installed on the road, and for this purpose, a vehicle detection sensor (Never-Fail Loop Sensor) installed on the road. And a vehicle detection controller (Loop Detection Controller) is provided on the side of the road, and a communication device for transmitting information detected by the vehicle detection sensor and a power supply device for supplying power should be configured.

Traffic information collection by the loop detector is a specific information is collected by the vehicle detection sensor installed on the road when the vehicle is traveling on the road, the collected information is analyzed and processed by the vehicle detection controller in real time to the traffic control center Is sent and performed.

Thirdly, the method through the image detector is that an image detector for analyzing the image collected by the surveillance camera is installed on the side of the road, and the analyzed information from the image detector is transmitted in real time to the traffic control center.

Finally, the collection method using a probe vehicle is a GPS satellite that collects information such as the vehicle location when driving a certain road with a traffic information collection terminal mounted on a commercial vehicle such as a taxi or a bus and a general vehicle. The vehicle receives the vehicle information and transmits the vehicle speed to the traffic control center, or collects traffic information by transmitting the driving speed for a predetermined distance to the traffic control center through a beacon installed on the road.

However, the traffic information collection method described above has the following problems.

First of all, the traffic information collected by the surveillance camera is expensive, including the surveillance camera, and the collected information is transmitted to the traffic control center through the optical cable, which is difficult to install on all roads. It was difficult to collect traffic information on the road.

In addition, since an industrial computer for image signal processing is required for each surveillance camera in order to measure the moving speed of the vehicle, the installation cost increases.

In addition, the roof detection method is difficult to install the vehicle detection sensor due to the excessive investment and burial work due to the excavation of the road, and the maintenance of the sensor damage due to frequent road construction and road damage. In addition, it could not be installed in the area where public network is not installed, which caused limitation of traffic information collection.

In addition, the collecting method through the image detector can collect traffic information at a lower cost than the above-mentioned collecting method, but it does not provide real-time video information or recognition of accidents such as traffic accidents, and separates the camera and the system from the industrial PC. Because it is operated, there is a problem in that the system construction cost is high and the power consumption is high and the system is frequently stopped. In addition, it could not be installed in areas where public networks were not installed, such as loop detectors.

Finally, the collection method using the probe vehicle can collect only the speed of the vehicle, and cannot collect traffic information when the vehicle is stopped. In addition, in order to collect more accurate and efficient traffic information, a lot of probe vehicles have to be put in, thus increasing the system construction cost and having low reliability of the collected information so that it cannot be linked with other traffic control systems such as signal linkage. there was. Of course, when installing beacons on the road, there were a lot of beacon installation costs and maintenance costs.

In other words, the above-described traffic collecting devices are configured separately from the system for analyzing and processing the actual collected information such as a camera, a loop detector and an image detector for collecting road information when collecting and transmitting road traffic information to a traffic control frame. It was difficult to install on the road, and because the information analysis is performed by the industrial PC base, a lot of difficulties such as initial system construction cost are consumed and maintenance is cumbersome.

In addition, the current traffic information camera provides only a simple function of collecting and analyzing only road traffic information, and there are no cameras capable of performing additional functions.

Accordingly, the present invention has been made to solve the above problems, and based on the video camera for imaging the image information of the vehicle running on the road and the image information captured from the video camera, traffic information such as traffic volume, speed, occupancy rate, etc. It is an object of the present invention to provide a network camera device having a traffic information collection and an additional function, and a traffic information processing method thereof, which integrates and miniaturizes an image detector to be analyzed to facilitate installation and maintenance.

In addition, the present invention not only manages traffic information generated on the road, but also stores image information on traffic accidents, provides traffic information for a certain region and climate information about the road, and recognizes whether the vehicle is being arranged. It has a different purpose to allow various additional functions to be performed.

In addition, the present invention has another object to transmit the collected traffic information to the traffic control center by selectively using a wireless LAN, a wired LAN and a mobile communication network according to the communication network conditions of the road where the network camera device is installed.

The network camera device having the traffic information collection and additional functions of the present invention for achieving the above object is configured to collect various traffic information generated on the road and transmit it to a vehicle detection system server (VDS Server) of the traffic control center. An image detecting camera apparatus, comprising: a camera module unit having a video camera to capture an image of the road, a digital signal processor for analyzing traffic information captured by the camera module unit, and collecting traffic information from the digital signal processor; Network processing unit for receiving and compressing the traffic information and real-time video information from the camera module unit to a vehicle detection system server (VDS Server) of the traffic control center by a predetermined communication method, the traffic information transmitted from the digital signal processing unit To detect and store accidents by analyzing them The digital signal processing unit analyzes the situation processing unit, the traffic information transmitting unit for transmitting the traffic information analyzed by the digital signal processing unit to the vehicle located in the predetermined area, and the mobile communication terminal of the vehicle driver located in the predetermined area. Text service execution unit for providing the traffic information is provided in the form of a short text service by the location-based service (LBS), Climate sensor interface unit that can be equipped with various sensors to detect the weather conditions for the road, The camera module unit A vehicle identification unit for capturing only the vehicle number image from the image information captured from the network and informing the case via the network processing unit according to the comparison result with the pre-stored arrangement, the digital signal processing unit and the network processing unit Power supply unit for supplying, and the yaw It is characterized in that the microprocessor for controlling the function of the cattle as a whole is configured integrally.

In addition, the traffic information processing method through the network camera device having the traffic information collection and additional functions of the present invention for achieving the above object, the process of setting the detection area for the traffic scene screen, the line / The vehicle target is extracted by a hybrid extraction technique based on a region segment, the tracking and recognition of the extracted vehicle target is performed, and the traffic information extracted from the tracking and recognition of the vehicle target is transmitted from the camera module unit. It is characterized in that the image information is transmitted to the traffic control frame made of a process that is analyzed.

Hereinafter, exemplary embodiments of the present invention will be described with reference to the accompanying drawings.

1 is a block diagram of a network camera apparatus having traffic information collection and additional functions according to a preferred embodiment of the present invention; FIG. 2 is a block diagram showing in detail the digital signal processor and the network processor of FIG. 2 is a detailed block diagram of the arrangement of the vehicle identification unit of FIG. 2.

As illustrated in FIG. 1, a camera module unit 102 equipped with a video camera for capturing road images and image information captured by the camera module unit 102 are analyzed for traffic volume, speed, occupancy rate, and atmosphere per lane. The digital signal processor 110 processes the traffic information such as a matrix length, an accident, such as a traffic accident, and the like, and receives the traffic information from the digital signal processor 110 and the real time video information from the camera module 102. And detects a sudden situation from the traffic information transmitted from the network processing unit 120 and the digital signal processing unit 110 to the vehicle detection system server (VDS Server) of the traffic control center by a predetermined communication method and the storage medium. The sudden situation processing unit 150 stored in the 150a, and the traffic information analyzed by the digital signal processing unit 110 to deliver to the vehicle located in a certain area The traffic information analyzed by the digital signal processing unit 110 is provided in the form of a short text service by the LBS to the traffic information transmitter 160 and the mobile communication terminal of the vehicle driver located in the predetermined area. Character service execution unit 170, the climate sensor interface unit 180 that can be equipped with a variety of sensors to detect the weather conditions for the road, the vehicle number image from the image information captured from the camera module unit 102 Arrangement vehicle recognition unit 190 for notifying through the network processing unit 120 in the case of a wanted vehicle according to the comparison result with the pre-stored wanted arrangement vehicle, the microprocessor 101 for overall control of the functions of the elements It is provided and configured integrally.

As shown in FIG. 2, the digital signal processing unit 110 includes a digital signal control unit 112 including a communication information collection algorithm for collecting traffic information based on the image information captured by the camera module unit 102. ), A flash memory (SDRAM) 114 for temporarily storing the image information at a predetermined address, and a flash ROM 114 for storing a program associated with driving the digital signal controller 112. . Here, the digital signal controller 112 employs a Philips 'TN1300 32-bit DSP chip'.

The traffic information collection algorithm includes a road image geometric transformation technique for setting a detection area S from an image screen, a vehicle target extraction algorithm using a vehicle target extraction algorithm, and a vehicle tracking and recognition algorithm. A vehicle tracking algorithm is provided for recognizing traffic information and accidents from the extracted vehicle target.

Subsequently, the network processing unit 120 compresses the traffic information collected by the digital signal processing unit 110 and the moving image information captured by the camera module unit 102 in a predetermined manner, for example, an M-JPEG method, and communicates with the communication module. The network control unit 122 and the video information are temporarily transmitted to the vehicle detection system server of the traffic control center 200 through any one of a wireless LAN, a wired LAN, and a mobile communication network of the unit 130. SDRAM 124 to be stored as a flash memory, and a flash ROM 126 to store a program related to network driving.

In addition, the network processing unit 120 is provided with a wireless LAN module 132, a wired LAN module 134 and a mobile communication module 136 so that information is transmitted through the wireless LAN, a wired LAN, and a mobile communication network.

Here, the network controller 122 is applied to the ANCP-I chip, which is a 32-bit RSIC processor for a network camera. The ANCP-I chip is equipped with an embedded Linux OS to manage overall control operations and all functions and to efficiently manage the system, and includes an ARM core, various device drivers, and a compression encoder. In addition, an application program such as a web server, a viewer software, and an Internet protocol may be provided.

On the other hand, the above-described configuration means, that is, the camera module unit 102, the digital signal processing unit 110, the network processing unit 120, the power supply unit 140 for supplying a predetermined level of power, and the unexpected situation processing unit 150 And a network camera device including a traffic information transmitting unit 160, a text service executing unit 170, a climate sensor interface unit 180, and a wanted vehicle recognition unit 190 compensate for humidity, temperature, and the like. It is designed to be protected from impact and wrapped in a cover member such as a camera housing so that it can be easily mounted on any structural bracket on the road.

Here, the camera module unit 102 mounted on the network camera device of the present invention is equipped with a thermal imaging camera that extracts thermal image information by measuring infrared radiation from not only a general CCTV but also a vehicle to be measured and road information as a background. Can be.

When the thermal imaging camera is applied, vehicle and road information can be accurately extracted even in a foggy or rainy weather or at night regardless of the climate. For example, an object with all temperatures emits an infrared ray with a constant wavelength on its own, while an infrared ray with a wavelength of about 8-12 um is emitted from an object with a low temperature (atmosphere temperature, human body temperature ....). Infrared rays having a wavelength of about 3-5 um are emitted from objects having a high temperature (engines, heaters, .....), and thus, vehicle information is obtained using such extracted infrared rays.

In addition, the wanted vehicle recognition unit 190 shown in FIG. 3 has an image recognition unit 190a for capturing only a vehicle number image from the image information captured by the camera module unit 102, and has vehicle number data for the wanted vehicle. A search unit 190c for accessing a vehicle number stored in the data storage unit 190b and the data storage unit 190b and searching for whether a vehicle number captured by the image recognition unit 190a exists, and according to the search result. When there is an arrangement, the vehicle information is transmitted to the vehicle detection system server 200 via the network processing unit 120, a transmission unit 190d.

5 is a diagram illustrating an embodiment in which a network camera device of the present invention is installed on a road.

As shown in the drawing, the network camera apparatus of the present invention can be easily installed at a road location for collecting traffic information because the video camera, the image detecting means and the network means are integrally installed.

The operation of the present invention configured as described above will be described with reference to FIGS. 1 to 9.

First, as shown in FIG. 6, the present invention provides a process of setting a detection area S for a traffic scene screen (ST100, ST102), and extracting a vehicle target from the set detection area S (ST104). ), The process of tracking and recognizing the extracted vehicle target (ST106), and the traffic information extracted according to the tracking and recognition of the vehicle target and the image information transmitted from the camera module unit 102 are traffic control center 200 ) And a process (ST110) in which traffic information is analyzed and provided.

The above process will be described in more detail with reference to FIGS. 7 to 9.

First, in order to collect traffic information by the network camera apparatus of the present invention, the digital signal processing unit 110 converts the 3D real world coordinate system into a computer image coordinate system by a camera correction operation for estimating a geometric relationship to a real world traffic scene. Conversion should be done.

Here, the camera correction operation is to estimate a parameter defining a relationship between a 3D space coordinate system and a 2D camera coordinate system, and is defined as an operation of extracting a visual variable of a camera located in an actual 3D space.

The camera visual variable includes an intrinsic parameter consisting of a focal length, a scale factor, an image center, and a lens distortion ratio, a pan angle, and a pan angle. Although it is classified into an extrinsic parameter consisting of a tilt angle and a swing angle, in the present invention, the focal length and the focal length of the visual variable are automatically set in order to automatically set the detection area S from the traffic scene screen. Only the pan and tilt angles will be used.

The detection area information S is extracted with respect to the current traffic scene screen captured by the camera module unit 102 using a specific correction pattern provided with only the focal length, the pan angle, and the tilt angle.

This is described as follows.

7 is a flowchart for setting a detection area by converting a real world information coordinate system to an image coordinate system according to the present invention.

First, when a predetermined power is supplied and driven from the power supply unit 140 in a state where the network camera device 100 of the present invention is mounted at a suitable location on a road, the camera module unit 102 may have a road width and a length. The three-dimensional real-world information coordinate system (Xw, Yw, Zw) for the traffic scene is captured and transmitted to the digital signal controller 112.

Then, the digital signal controller 112 converts the three-dimensional real world information coordinate system (Xw, Yw, Zw) by a rotation matrix and a translation vector by a traffic information collection algorithm. (x, y, z) and calculate the pan angle and tilt angle of the camera by the automatic camera adjustment function (ST120, ST122).

The camera coordinate system (x, y, z) is calculated by the following [Formula 1].

---[식 1]

--- [Equation 1]

Here, 'R' is a rotation matrix, and 'T' is defined as follows as a motion vector.

On the other hand, the pan angle and tilt angle of the camera are calculated by the automatic camera adjustment function, but the focal length is set to the automatic camera adjustment function in a state where the real world plane and the CCD sensor coordinate system of the camera are symmetrically converted to each other about the optical axis Zw. Calculated by camera calibration.

As such, when the focal length is calculated, the coordinate system (xu, zu), which becomes the actual camera image coordinate system, is calculated by the perspective transformation method of the following [Equation 2] (ST124).

xu = f (x / y)

zu = z / y ------ [Equation 2]

When the camera image coordinate system (xu, zu) is extracted by [Equation 2], it is converted into image coordinate systems (xf, zf) which are then processed by a real computer, and the image coordinate system (xf, zf) is the CCD length of the set X axis. And the Y-axis CCD length and actual image size (ST126).

Therefore, referring to FIG. 4, road image information (FIG. 4A) is converted into detection area information (FIG. 4B) S by the geometric transformation technique captured by the actual camera module unit 102 (ST128). ).

On the other hand, the vehicle target is extracted by the vehicle tracking algorithm mounted on the digital signal controller 112, that is, the Philips 'TN1300 32-bit DSP chip' from the detection area S information of the road image geometrically converted by the above-described method. Traffic information such as vehicle speed and type of vehicle is recognized.

This will be described with reference to FIG. 8, which is a flowchart for extracting a vehicle target.

First, a method of extracting a target of the vehicle includes a method of extracting a background differencing based on a difference between background image information and input image information, a method of extracting only a moving target by obtaining a difference of input image information, There is a method of extracting by a difference in brightness between a background image and a target (space differencing) and a method of extracting a target with statistics of elements constituting the background image (Gray-level comparison). A new vehicle target extraction algorithm will be provided by combining only the background differencing by the difference from the image and the space differencing by the brightness difference between the background image and the target.

The vehicle target extraction algorithm is a hybrid target extraction technique using a line / area segment, which detects a predetermined number of lines of detection area S set in each lane or road in the image screen. In operation S130, the area is divided to recognize the shape of the vehicle.

Subsequently, the background image and the input image information are compared with each other in the segmented line and area, and then the brightness information, the edge information of the target, the length / width information of the target, pixel ratio, average value, or the like. The feature of the target is extracted by the standard horseshoe value or the like (ST132, ST134).

In other words, in the state in which the background image information, which is a reference, is stored in the SDRAM 114 based on the detection area S information set by the digital signal controller 111 as shown in FIG. 4B. The difference between the image information input from the unit 102 and the background image information stored in the SDRAM 114 is extracted by the digital signal controller 112 to determine whether a vehicle exists on the road.

Accordingly, the vehicle target extraction algorithm is activated in accordance with the control operation of the digital signal controller 112 to extract the vehicle target, and then the vehicle tracking for recognizing traffic information and accidents based on the extracted vehicle target. Algorithm and vehicle recognition algorithm are performed.

FIG. 9 is a flowchart illustrating tracking and recognizing a vehicle target extracted by FIG. 8.

First, in order to track and recognize the extracted vehicle target by the digital signal processor 110, one or more vehicle targets extracted from road image information and background image information should be input (ST140).

Subsequently, according to the vehicle tracking algorithm, the template matching method and the 'Nearest Neighbor Assoctation' method according to the correlation method described in [Equation 3] are applied to track the vehicle in the line and area in the detection area S. Is performed.

-------식[ 3]

------- Equation [3]

Where s = 0,1,2,... , M-1, t = 0,1,2,... , N-1.

In addition, w is the pixel mean value of w (xy), f (x, y) is the pixel mean value of (x, y) corresponding to the w (x, y) position region, and the denominator of r (s, t) is w ( Normalized by the average of x, y and f (x, y).

In addition, the correlation index r (s, t) is in the range -1 ≤ r ≤ 1, and has an independent value regardless of the image amplification of f (x, y) and w (x, y).

Geometrical transformation is performed by the above equation (3), and tracking of the vehicle target extracted from the line and area segments is performed (ST142, ST144).

At this time, in order to track the target of the vehicle successfully, the target trajectory management is an important factor.

Table 1 shows an example of the trajectory management.

TABLE 1

In [Table 1], since the score increases according to how long the trajectory lasts, the trajectory disappears when the trajectory score is low and less than the set score when a specific target is not detected. That is, the vehicle B and the vehicle C are partially overlapped with the position of the vehicle A.

Accordingly, the locus management can check whether the locus number and the target overlap, loss or the like, indicating a sudden situation such as a traffic accident by obtaining information on the position change of the two targets at the current location.

The association between the measurement and the trajectory for the target management is calculated by the Nearest Neighbor Assoctation algorithm.

Meanwhile, when vehicle tracking is normally performed by the vehicle tracking algorithm and the vehicle recognition algorithm, traffic information may be calculated or a sudden situation may be recognized by the above-described movement trajectory of the vehicle (ST146).

The traffic information is calculated by vehicle tracking recognition according to the line segment, and the sudden situation recognition is performed by vehicle tracking recognition according to the area segment.

That is, the vehicle tracking recognition according to the line segment may be generated after first selecting a plurality of line tracking patterns from the initial frame of the detection area S captured by the camera module unit 102 and selecting the line tracking patterns. From the frame, each line tracking pattern selected in the detection area S is traced. At this time, the number of frames is counted while the line tracking pattern is traced, the distance is corrected by counting the number of pixels in the last frame, and the vehicle signal and occupancy rate in the current detection area S are calculated by the digital signal controller 112.

In this case, in order to reduce the error of the vehicle speed, the average value may be calculated using only the remaining vehicle pattern except for the maximum and minimum moving vehicle patterns in each line tracking pattern.

In addition, the vehicle can be tracked according to the area segment, which can be recognized in the detection area S by extracting an object in the form of a vehicle for each frame. If the trajectory of the vehicle object is stopped or overlapped in the middle by the track management during the vehicle tracking recognition, the digital signal controller 112 recognizes the accident as an accident such as a traffic accident, and the accident is tracked according to the line segment. It is processed in conjunction with the traffic information calculated from the recognition.

As such, the digital processing control unit 112 collects traffic information by a vehicle tracking algorithm, and the collected traffic information is sent to the network processing unit 120.

Then, the network processing unit 120 detects the vehicle of the traffic control center 200 through the communication network currently installed on the road to the traffic information and image information applied from the camera module unit 102 and compressed by the M-JPEG method. Send it to the VDS Server.

That is, the network processing unit 120 is a mobile communication with a wireless LAN module 132 and a wired LAN module 134 equipped with a predetermined protocol so as to transmit the traffic information and image information through a wireless LAN, a wired LAN, and a mobile communication network. Since the module 136 is mounted, data can be transmitted through the most suitable communication medium.

Thereafter, the vehicle detection system server of the traffic control center 200 receives the traffic information and the image information, and analyzes the traffic information while displaying it through a display such as a monitor.

The analyzed traffic information is provided to an unspecified number through the Internet network, the public network, or the cable network, to control the traffic signal, or to the support agency communication network so as to quickly deal with accidents.

In addition, after the embedded Linux OS and other device drivers are installed on the ANCP-I chip, a 32-bit RSIC processor for network cameras, applications such as a web server, viewer software, and an Internet protocol can be installed.

In addition, when the detection area S for the traffic scene screen is set by the digital signal processing unit 110, a program for adjusting the detection area S may be driven to verify and optimize the traffic information collecting algorithm on a PC. When applicable.

Meanwhile, various additional functions are supported in the network camera device of the present invention, which will be described in more detail.

First, a sudden situation processing unit 150 is provided to detect a sudden situation such as a traffic accident and store image information thereof. Here, the detailed description of the incident situation will be omitted.

The abrupt situation processing unit 150 is an abnormality such as overlapping or losing the target by the vehicle tracking recognition according to the above-described area segment while the road corresponding to the predetermined area range is continuously imaged by the camera module unit 102. If it is confirmed, it is recognized that an accident such as a traffic accident has occurred.

Then, the sudden situation processing unit 150 detects the upper road image information on the time point at which the abnormality occurs and stores the image information on the time point in the storage medium 150a.

The stored image information is provided as information for traffic accident processing, and in some cases, image information can be provided in association with an insurance company.

Secondly, the traffic information transmitting unit 160 is provided to provide traffic information for the corresponding area through the FM frequency of the vehicle operating within a predetermined distance from the location where the network camera device of the present invention is mounted.

For example, when a vehicle is running near 300 km from the downhill line of Seoul at the Gyeongbu Expressway with a plurality of network cameras installed at a predetermined distance, the network camera installed at the location where the vehicle is traveling is the digital signal processor 110. Traffic information about the neighboring area received by the network processing unit 120 from the traffic information and the traffic control center 200 near the starting point of the Gyeongbu Expressway downline Seoul collected by the It can be heard through the FM frequency.

In addition, when the vehicle passes near the 300 km and passes the area where the other network camera is installed, the vehicle can listen to the traffic information about the area from the corresponding network camera.

Of course, the FM frequency should be well known to drivers before, and in addition to the FM frequency, a transmission means is provided in the network camera and a reception means for receiving traffic information from the transmission means may be separately installed. The transmitting and receiving means may be a device such as a wireless LAN. In this case, the network camera is provided with a transmitting WLAN, and a receiving WLAN for receiving a signal from the transmitting WLAN must be mounted in a vehicle. do. This is different from traffic broadcasting broadcasted through a radio broadcasting network nationwide or locally.

That is, the vehicle driver can listen to the road traffic information on the road where he is actually located from the network camera in real time.

Third, the character service function by the character service execution unit 170 is to determine the location of the terminal of the driver of the vehicle by the location-based service of the mobile communication network when the vehicle is driving around 300km from the Seoul starting line down the Gyeongbu Expressway, It is a text data service of traffic information of the road collected from the network camera installed at the nearest point to the point.

The road traffic information collected and analyzed by the network camera is transmitted to the traffic control center 200 through the network processing unit 120, and the transmitted road traffic information is transmitted to the mobile communication network 300 again. The communication network 300 is provided by providing the road traffic information to a mobile communication terminal of a vehicle driver driving the road in the form of short text.

Of course, the text service function is performed according to a separate service subscription.

At this time, the text service execution unit 170 is executed through the mobile communication network 300 as described above, the text service execution unit 170 is provided in the traffic information transmitting unit 160 integrally, traffic control When the road traffic information is transmitted from the center 200 to the traffic information transmitting unit 160, the traffic information transmitting unit 160 may provide the road traffic information to the mobile communication terminal of the vehicle driver according to a text service function. .

Fourth, the climate sensor interface unit 180 is a variety of sensors that can detect the climate change around the road, that is, snow, rain, fog, wind speed, temperature .... Optionally mount sensors to detect the back.

For example, when the corresponding sensor is installed in the climate sensor interface unit 180 to recognize the rainfall state, it is possible to know in which region the current rain is falling by the sensor, and the weather information as described above. The network processor 120 may provide the vehicle driver with the vehicle driver.

The climate sensor interface unit 180 may be optionally mounted with one or more, and it is preferable to mount an optimal sensor for sensing a climate that can occur most frequently on any road. In other words, bridges with a lot of wind compared to other areas should continuously detect wind strength and predict wind gusts. The climate sensor interface unit 180 of the network camera installed in the bridge should be equipped with a wind speed sensor. For example, sensors that detect the occurrence of fog should be installed in areas where fog occurs frequently.

However, in addition to selectively coupling and removing each sensor to the climate sensor interface unit 180, an optimal sensor capable of recognizing major climate change in a specific region as described above may be integrally provided. Of course. That is, in the area where the fog is frequently generated, a sensor for detecting whether fog is generated is not assembled to the climate sensor interface unit 180, but the sensor may be integrally installed.

Fifth, the arrangement vehicle recognition unit 190 enables the user to quickly recognize whether the arrangement is for all vehicles traveling on the road where the network camera of the present invention is installed.

It captures the vehicle number image from the image information captured by the camera module unit 102 by the image recognition unit 190a, and the data storage unit 190b having the vehicle number data for the arrangement by the search unit 190c Is accessed and the vehicle number captured by the image recognition unit 190a exists.

And, if there is an arrangement vehicle according to the search result to transmit the vehicle information to the traffic control center 200 via the network processing unit 120, the traffic control center 200 is a police terminal ( 400 and the like, so that the arrested vehicle information may be transmitted in real time.

At this time, the data storage unit 190b is linked to the police-related terminal 400 to update and store the arrangement vehicle data at predetermined intervals.

As described above, according to the network camera apparatus having the traffic information collection and additional functions of the present invention and the traffic information processing method thereof, information on the road and various traffic information for driving the road, such as traffic volume, vehicle speed, road occupancy rate, Integrate and miniaturize the camera and video detection device that can collect and analyze traffic information such as traffic accidents, and the wired / wireless communication module that can transmit the collected information to the traffic control center through various network interfaces. It can be easily installed on the road and the system construction cost can be reduced.

In other words, the traffic collection information through the conventional CCTV, image detector, loop detector, and probe vehicle is very expensive to install and operate because the camera and image detector are operated on an industrial PC to recognize and analyze the image information of the road on the side of the road. It was difficult to carry and maintain, and it could not be installed in areas without public network, so the collection of traffic information was not efficient.

In addition, the network camera device of the present invention is expected to collect various traffic information which cannot be partially collected by the conventional traffic collection device.

Claims (19)

In the video detection camera device configured to collect various traffic information generated on the road and transmit to the vehicle detection system server (VDS Server) of the traffic control center, Camera module unit provided with a video camera to capture the image of the road, A traffic information collection algorithm is included, which includes a process for recognizing a vehicle based on image information photographed on the road, analyzing the movement state of the recognized vehicle, and collecting traffic information. A digital signal processor which analyzes the image information photographed from the camera module unit and collects traffic information such as traffic volume, speed, occupancy rate, queue length, accidental situation such as accident / failure, etc. A network processor for receiving and compressing traffic information from the digital signal processor and real-time video information from the camera module unit and transmitting the compressed traffic information to a vehicle detection system server (VDS server) of the traffic control center by a predetermined communication method; A sudden situation processor for recognizing a sudden situation according to the traffic information transmitted by the digital signal processor and storing image information accordingly; The traffic information transmitter transmits the traffic information analyzed by the digital signal processor to a vehicle located in a predetermined area and the traffic information analyzed by the digital signal processor to a mobile communication terminal of a vehicle driver located in the predetermined area. Character service execution unit to be provided in the form of a short text service by the location-based services (LBS), Climate sensor interface unit that can be equipped with a variety of sensors to detect the weather conditions on the road, Arrangement vehicle recognition unit for capturing only the vehicle number image from the image information captured by the camera module unit and notifies through the network processing unit in the case of the arrangement vehicle according to the comparison result with the arrangement vehicle already stored; A power supply unit supplying constant power to the digital signal processor and the network processor; And a microprocessor for controlling the functions of the elements as a whole. The method of claim 1, The digital signal processor; A digital signal controller incorporating the traffic information collection algorithm for collecting traffic information based on the image information captured by the camera module unit, a flash ROM for storing a program related to driving the digital signal controller, A network camera device having traffic information collection and additional functions, comprising: an SDRAM for temporarily storing the image information at a predetermined address. The method according to claim 1 or 2, The traffic information collection algorithm is; The vehicle image is extracted and correlated by the road image geometric transformation technique for setting the detection region, which is a computer image coordinate system, from the image screen, and the vehicle target extraction algorithm to which the hybrid extraction technique of the line / region segment is applied. And a vehicle tracking algorithm for recognizing traffic information and accidents from the extracted vehicle target by an applied vehicle tracking and recognition algorithm. The method of claim 2, The digital signal controller; Philips 'TN1300 32-bit DSP chip' is a network camera device with traffic information collection and additional features. The method of claim 1, The network processor; Compress the traffic information collected by the digital signal processing unit and the video information captured by the camera module unit according to a predetermined method, and transmit it to the vehicle detection system server through any one of a wireless LAN, a wired LAN, and a mobile communication network. And a network control unit for transmitting the traffic information, wherein the network camera apparatus has additional functions. The method of claim 5, The network control unit; A network camera device having traffic information collection and additional functions, which is an 'ANCP-I' chip that is a 32-bit RSIC processor for network cameras. According to claim 3, wherein the digital signal processing unit in the built-in traffic signal collection algorithm, the sudden situation determination continuously analyzes the image information captured from the camera module unit, the overlapping and loss of the vehicle target by a track management or a fixed time A network camera device having traffic information collection and additional functions, which is determined to be a sudden situation such as a traffic accident or a vehicle failure when there is no abnormal mobile situation. The method of claim 1, The traffic information sending unit; Traffic information collection and additional functions, characterized in that the traffic information for the specific area collected by the digital signal processing unit or the traffic information for the area where the vehicle is located from the vehicle detection system server is transmitted to the FM receiver of the vehicle. Network camera device having a. The method of claim 8, The traffic information sending unit; Traffic information collection and addition, characterized in that the transmission means for transmitting traffic information to the receiving means may be further configured in a state in which a receiving means for receiving traffic information within a predetermined range is separately installed in the vehicle. Network camera device with the function. The method according to claim 1 or 8, The text service execution unit; The road traffic information collected and analyzed from the network camera is provided from the traffic control center to a mobile communication terminal of a vehicle driver driving the road in the form of short texts via a mobile communication network, or integrated into the traffic information transmitting unit. The network camera device having traffic information collection and additional functions, the road traffic information received from the traffic control center being provided to the mobile communication terminal. The method of claim 1, The climate sensor interface unit; Network camera device having a traffic information collection and additional functions, characterized in that the various weather detection sensors are detachable to detect the road and the climate around the road. The method of claim 11, The climate sensor interface unit; At least one climate sensor among the climate sensor according to the weather conditions of the road network camera device having a traffic information collection and additional functions, characterized in that can be configured integrally provided. The method of claim 1, The arrangement vehicle recognition unit; Image recognition unit for capturing only the vehicle number image from the image information captured from the camera module unit, A data storage unit having vehicle number data for an arrangement vehicle; A search unit for accessing the vehicle number stored in the data storage unit and searching whether the vehicle number captured by the image recognition unit exists; And a transmission unit for transmitting the vehicle information to the vehicle detection system server through the network processing unit when there is an arrangement vehicle according to the search result. The method of claim 1, The camera module unit may be equipped with a thermal image camera to extract thermal image information by measuring infrared radiation from vehicle information to be measured and road information as a background regardless of the climate, and additional information. Network camera device having a. A video camera for capturing image information of the road and an image detector for recognizing and analyzing the collected image information are integrally formed to collect traffic information and transmit it to a vehicle detection system server (VDS server) of the traffic control center. In the method for analyzing and providing, Extracting camera visual variables and converting them into 2D computer images for 3D traffic scenes according to the extracted visual variables In the set detection area, detection area information set for each lane or road in the image information is divided into predetermined lines and areas according to a hybrid extraction method by line / area segments, and the background image information Extracting a vehicle target by extracting a target feature of the vehicle by comparing previously input image information; Tracking and recognition of the extracted vehicle target, The network camera device having traffic information collection and additional functions, wherein the traffic information extracted from the tracking and recognition of the vehicle target and the image information transmitted from the camera module unit are transmitted to the traffic control frame for analysis. Traffic Information Processing Method The method of claim 15, The detection area setting process is; The three-dimensional real world information coordinate system (Xw, Yw, Zw) of the traffic scene screen is converted into a three-dimensional camera coordinate system (x, y, z) by a rotation matrix and a translation vector. And calculating the pan angle and tilt angle of the camera by the automatic camera adjustment function. Extracting the camera coordinate system (x, y, z) into an image coordinate system (xu, zu) by a perspective perspective method having a pinhole lens geometric transformation, The image coordinate system (xu, zu) is extracted to the computer image coordinate system (xf, zf) by the CCD length of the X-axis, the CCD length of the Y-axis and the actual image size has a traffic information collection and having additional functions Traffic information processing method of a network camera device. delete The method of claim 15, The vehicle target extraction, Network camera device having additional information and traffic information collection, characterized in that the difference between the brightness information, the target edge information, the target length / width information, pixel ratio, average value or standard deviation value between the background image and the input image Traffic Information Processing Method The method of claim 15, Vehicle tracking for the extracted vehicle target is performed by vehicle tracking of line segment and area segment, and traffic information of vehicle speed and road occupancy are calculated by the line segment, and overlapping of vehicle trajectory is generated by the area segment. A traffic information processing method for a network camera device having traffic information collection and additional functions, wherein an accident situation is recognized based on a stationary state.

Images