KR20120126764A - A vehicle image detecting system and the method thereof - Google Patents

Abstract

PURPOSE: A vehicle image sensing system and a method thereof are provided to easily calculate a speed of a vehicle by using distance and image frame time by sensing a predetermined starting point and an end point. CONSTITUTION: A vehicle image detection system senses a first point and a second point of a vehicle which passes a predetermined area(110). The system obtains a passage video from the first point to the second point(120). The system calculates speed of the vehicle by using distance between the first point and the second point(130). [Reference numerals] (110) Sensing a first point and a second point of a vehicle which passes a predetermined area; (120) Obtaining a passage video from the first point to the second point; (130) Calculating speed of the vehicle by using distance between the first point and the second point; (AA) Start; (BB) End

Description

Vehicle image detection system and method thereof {A VEHICLE IMAGE DETECTING SYSTEM AND THE METHOD THEREOF}

The present invention relates to a vehicle image detection system and a method thereof, and more particularly, to a vehicle image detection system and method that can easily obtain the image information of the speed and license plate of a moving vehicle.

As the social structure becomes more complex and diversified, the area of activity of the individual gradually expands, and as the rapid movement is required, the use of automobiles has increased to facilitate this. In addition, due to the development of logistics, the number of vehicles for delivery is rapidly increased.

Nowadays, as the social and economic scale expands and becomes more advanced, the demand for vehicles and transportation has exploded, which has led to serious transportation problems in large cities.

In addition, as the use of vehicles is generalized, traffic accidents caused by vehicles are increasing, and such traffic accidents cause not only physical damage but also fatal human damage.

In order to prevent such traffic accidents, road speed bumps may be installed as a method for preventing high speeds in an accident-prone area or a dangerous area, but such speed bumps may be used only for automobiles, highways and highways. As such, it cannot be installed in a section in which the vehicle travels at high speed.

Therefore, a method for preventing overspeed of a vehicle by using a speeding vehicle on a road has been used. For example, a loop sensor is provided on a specific road at a predetermined interval, and whether the vehicle is speeded by measuring a vehicle speed with a time difference and a distance value detected by the loop sensors, and the identified vehicle is separately There is a method of taking a picture using a camera.

However, in the prior art, since the roof-type sensor must be installed on the road, the installation cost is high, and the maintenance cost is increased due to failure due to road breakage or the like when used for a long time.

In addition, since the camera operating in conjunction with the loop-type sensor must be installed separately, this also increased the cost.

Therefore, there is a need for a system and method for simply obtaining image information of a speed plate and a license plate of a moving vehicle.

The present invention provides a vehicle image detection system and method for easily calculating a vehicle speed by using a distance and an image frame time by acquiring a moving image of a vehicle moving in a set section and detecting a predetermined starting point and an ending point.

The present invention provides a vehicle image detection system and method capable of confirming both the front part and the rear part of a vehicle by providing a moving image passing through a start point and an end point.

The present invention provides a vehicle image detection system and method that can easily extract vehicle number information by acquiring a still image from a video acquired in a set section.

Vehicle image detection method according to an embodiment of the present invention, the first step of detecting the passage of the first point and the second point of the vehicle passing through the predetermined area, the passing video from the first point to the second point And a third step of calculating the speed of the vehicle using the number of image frames between the first point and the second point, the interframe time, and the distance between the first point and the second point. It includes.

According to an aspect of the present invention, the detecting of the first step comprises: generating a background image every frame by using a Gaussian mixture model, comparing the starting point area set at the top of the image with the current frame, and starting point at which the vehicle enters. And detecting the passing of the first point, and detecting the passing of the second point, which is an end point at which the vehicle disappears by comparing the end point area set at the bottom of the image with the current frame.

According to an aspect of the present invention, a fourth step of obtaining a still image for license plate recognition of the vehicle from the passing video from the first point to the second point, a fifth to recognize the license plate in the obtained license plate recognition still image (6) storing the video frame corresponding to the first point and the video frame corresponding to the second point as a video clip, and generating a compressed video clip by encoding the stored video clip through a predetermined video compression technique; and The method may further include transmitting the traffic information including the compressed video clip, the calculated speed information of the vehicle, and the recognized license plate information to an image management unit that performs image output and management.

According to an aspect of the present invention, the fourth step may be obtained as a still image for recognizing a license plate of the vehicle, the intermediate viewpoint still image of the passing video from the first point to the second point.

According to one aspect of the present invention, decoding the received compressed video clip, and outputting the traffic information including the vehicle number and speed from the image management unit 8 and step 9 to store the decoded video and traffic information in the database It may further include.

Vehicle image detection system according to an embodiment of the present invention, the passing point recognition unit for detecting the passage of the first point and the second point of the vehicle passing through the predetermined area, the passage from the first point to the second point An image acquisition unit for acquiring a moving image, a speed calculation for calculating a speed of a vehicle using an image frame number between the first point and the second point, an interframe time, and a distance between the first point and the second point And an image management unit for outputting and storing at least one of the obtained video and the speed of the vehicle.

According to an aspect of the present invention, a vehicle number extraction unit for obtaining a still image for license plate recognition of the vehicle from the passing video from the first point to the second point and recognizes the license plate in the obtained license plate recognition still image, the An encoding unit for storing the video frame corresponding to the first point and the video frame corresponding to the second point as a video clip, and generating a compressed video clip by encoding the stored video clip through a predetermined video compression technique. The apparatus may further include a transmitter configured to transmit the clip and the recognized license plate information to the image manager.

According to an aspect of the present invention, the image management unit, a decoding unit for decoding the received image, the output unit for outputting the traffic information including the vehicle number and speed and the decoded image and the traffic information to store It can include a database.

According to an embodiment of the present invention, by acquiring a video of a moving vehicle in a set section and by detecting a predetermined start point and end point, the vehicle image detection system that can easily calculate the speed of the vehicle using the distance and image frame time And a method are provided.

According to an embodiment of the present invention, by providing a video passing through the start point and the end point, there is provided a vehicle image detection system and method that can be confirmed from both the front portion and the rear portion of the vehicle.

According to an embodiment of the present invention, a vehicle image detection system and method for easily extracting vehicle number information by acquiring a still image from a video acquired in a set section is provided.

1 is a flowchart illustrating a vehicle image detecting method according to an embodiment of the present invention.
2 is a flowchart illustrating a method for detecting a passing point of a vehicle according to an exemplary embodiment of the present invention.
3 is a diagram illustrating a screen for detecting a start point and an end point according to an embodiment of the present invention.
4 is a flowchart illustrating a process of transmitting the extracted information to an image management center according to an embodiment of the present invention.
5 is a flowchart illustrating a vehicle image detection system according to an embodiment of the present invention.

Hereinafter, with reference to the contents described in the accompanying drawings will be described embodiments of the present invention; However, the present invention is not limited to or limited by the embodiments. Like reference symbols in the drawings denote like elements.

1 is a flowchart illustrating a vehicle image detecting method according to an embodiment of the present invention.

Referring to FIG. 1, in step 110, a passage of a first point (start point) and a second point (end point) of a vehicle passing through a preset area may be detected. Here, the first point is a point for determining whether the vehicle enters the designated area, and when the vehicle enters the first point, video recording may be started. In addition, the second point is a point for determining whether the vehicle exits the designated area, and when the vehicle passes the second point, video shooting may be terminated. Therefore, the speed of the moving vehicle can be easily calculated by detecting whether the preset start point and end point pass, passing time, and the like. Here, the process of detecting the passage of the first point and the second point will be described in detail below with reference to FIG. 2.

2 is a flowchart illustrating a method of detecting a passing point of a vehicle according to an exemplary embodiment of the present invention.

Referring to FIG. 2, in step 210, a background image may be generated every frame using a Gaussian mixture model. In this case, the start point region may be displayed at the top of the background image and the end point region may be displayed at the bottom of the background image to determine the passing of the starting point and the end point of the vehicle.

In operation 220, the passing of the first point of the vehicle may be detected by comparing the current frame with the starting point region set at the top of the image through the background image. For example, the difference image between the current frame and the background image may be calculated, and the passage of the first point of the vehicle may be detected using the difference image.

In operation 230, the passing of the second point of the vehicle may be detected by comparing the current frame with the end point region set at the bottom of the image through the background image. As described above, the difference image between the current frame and the background image may be calculated, and the passage of the second point of the vehicle may be detected using the difference image.

Referring back to FIG. 1, in operation 120, a passing video from the first point to the second point may be acquired. Here, the passing through video may start moving picture recording from the moment of detecting the passage of the first point of the vehicle and end the moving picture shooting at the moment of detecting the passage of the second point of the vehicle, thereby obtaining the passing video of the corresponding portion. In addition, as another example of obtaining a passing video, the moving picture may be set to be continuously photographed, and the moving picture between the first point passing point and the second point passing point of the vehicle may be separated from the continuously photographed video.

In operation 130, the speed of the vehicle may be calculated using the number of image frames between the first point and the second point, the interframe time, and the distance between the first point and the second point. Since the speed can be obtained by dividing the time by the distance, the speed of the vehicle can be calculated according to Equation 1 below.

For example, if the actual distance between the first point and the second point is 10m, the frame difference between the second point image frame and the first point image frame is 100 frames, and the time between frames is 1/60 second, the speed of the vehicle is The vehicle speed can be calculated at 6.02 m / s or 21.67 Km / h. Accordingly, the speed of the moving vehicle can be calculated more easily and accurately by using the number of image frames, the interframe time, and the distance between the first point and the second point between the first point and the second point.

3 is a diagram illustrating a screen for detecting a start point and an end point according to an embodiment of the present invention.

Referring to FIG. 3, an example screen in which a start point region 310 and an end point region 320 are set in a background image to detect whether a vehicle enters or exits a set region is illustrated. That is, when the vehicle enters the preset area, the specific part of the vehicle (eg, the front part of the vehicle) passes the starting point area 310, and at this point, it can be sensed that the vehicle has entered the preset area. In addition, when the vehicle exits from the preset area, the specific part of the vehicle (eg, the front part of the vehicle) passes through the endpoint area 320, and thus, it may be sensed that this is the time when the vehicle exits from the preset area. Here, since the actual distance between the start point area 310 and the end point area 320 does not change, the speed of the vehicle may be calculated by determining the time at which the vehicle moves from the start point area 310 to the end point area 320.

4 is a flowchart illustrating a process of transmitting the extracted information to the image management unit according to an embodiment of the present invention.

Referring to FIG. 4, in operation 410, a still image for recognizing a license plate of a vehicle may be acquired from a passing video from the first point to the second point. Here, in order to recognize the license plate of the vehicle, the part including the license plate must be accurately contained in the still image. Therefore, as an example for obtaining a still image for license plate recognition of the vehicle, a still image when the vehicle is located at the center of the first point and the second point may be obtained as the still image for license plate recognition. To this end, an intermediate view still image of the passing video from the first point to the second point can be obtained as a still image for recognizing the license plate of the vehicle, which is a function of the first time passing time and the second time passing time of the vehicle. A method of obtaining a still image corresponding to the intermediate value by obtaining an intermediate value may be used. The method for obtaining a still image for recognizing a license plate of a vehicle is not limited to the above-described embodiment, and various methods may be used.

 In operation 420, the license plate may be recognized from the acquired still image for license plate recognition. Here, various methods such as pattern matching and machine learning may be used as an image processing method for recognizing numbers or letters in still images.

In operation 430, the video frame corresponding to the first point and the video frame corresponding to the second point may be stored as a video clip, and the compressed video clip may be generated by encoding the stored video clip through a predetermined video compression technique. have. Here, the storage of the video clip may be stored as a video clip by adding a predetermined frame in the frame corresponding to the second point in consideration of the constant frame margin at which the vehicle exits from the image in the frame corresponding to the second point. Similarly, the video may be stored from a frame before a frame corresponding to the first point and stored in consideration of the frame margin.

In operation 440, the traffic information including the compressed video clip, the calculated speed information of the vehicle, and the recognized vehicle license plate information may be transmitted to an image management unit that performs image output and management.

In operation 450, the received compressed video clip may be decoded and traffic information including the vehicle number and speed may be output from the image manager.

In operation 460, the decoded video and the traffic information may be stored in a database.

As described above, the starting and ending points of the vehicle passing through the set area are sensed, and the speed of the moving vehicle can be easily and accurately calculated using the distance between the starting and ending points, the number of frames passing through the starting and ending points, and the time between frames. Can be.

In addition, the present invention provides a vehicle image detection system and method capable of detecting both the front part and the rear part of the vehicle by providing the entire video passing through the start point and the end point.

In addition, the license plate information can be easily extracted by calculating the intermediate time passing through the start point and the end point to obtain a still image corresponding to the intermediate time.

5 is a flowchart illustrating a vehicle image detection system 500 according to an embodiment of the present invention.

Referring to FIG. 5, the vehicle image detection system 500 includes a passing point recognizer 510, an image acquirer 520, a speed calculator 530, and an image manager 570.

The passing point recognition unit 510 may detect the passage of the first point and the second point of the vehicle passing through the preset area. Here, the first point is a start point at which the vehicle starts to enter the preset area, and the second point is an end point at which the vehicle leaves the preset area. In order to detect whether the vehicle passes the first and second points, a background image is generated every frame using a Gaussian mixture model, and the starting point area or the end point area and the current frame set at the top or bottom of the image through the background image. Compared to the first point or the second point of the vehicle can be detected.

The image acquisition unit 520 may acquire a passing video from the first point to the second point. That is, the moving point from the first point to the second point can be obtained by detecting a moving point between the first point and the second point of the vehicle and acquiring a moving picture therebetween.

The speed calculator 530 may calculate the speed of the vehicle using the number of image frames between the first point and the second point, the interframe time, and the distance between the first point and the second point. That is, the vehicle speed may be calculated by dividing the distance between the first point and the second point by multiplying the number of image frames between the first point and the second point by the interframe time.

In addition, the vehicle image detection system 500 may further include an encoder 540, a vehicle number extractor 550, and a transmitter 560.

The encoding unit 540 generates a compressed video clip by storing the video frame corresponding to the first point and the video frame corresponding to the second point as a video clip, and encoding the stored video clip through a predetermined video compression technique. can do.

The vehicle number extraction unit 550 may acquire a still image for license plate recognition of the vehicle from the passing video from the first point to the second point, and recognize the license plate in the obtained license plate recognition still image. At this time, the license plate recognition image of the vehicle should be an image including the entire license plate, which may be used to obtain a still image corresponding to the intermediate time between the first point and the second point passing time.

The transmitter 560 may transmit the compressed video clip and the recognized license plate information to the image manager 570. Here, the transmission of the video clip and license plate information may be made through wired or wireless communication.

The image manager 570 may output and store at least one of the obtained video and the speed of the vehicle. In addition, the image manager 570 may display and store the license plate information of the vehicle. Accordingly, the image management unit 570 may include a decoding unit 571 for decoding the received image, an output unit 572 for outputting the traffic information including the vehicle number and speed, and the decoded image and the traffic. It may include a database 573 that stores information.

As described above, by acquiring a moving image of the vehicle moving in the set section and detecting a preset start point and end point, the speed of the vehicle can be easily calculated using the time between the distance and the image frame.

In addition, by obtaining a still image including a license plate image of the video obtained in the set section, it is possible to easily extract the vehicle number information.

In addition, the vehicle image detection method according to an embodiment of the present invention includes a computer readable medium including program instructions for performing operations implemented by various computers. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. The media may be program instructions that are specially designed and constructed for the present invention or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like.

Although one embodiment of the present invention as described above has been described by a limited embodiment and drawings, one embodiment of the present invention is not limited to the above-described embodiment, which is a general knowledge in the field of the present invention Those having a variety of modifications and variations are possible from these descriptions. Accordingly, one embodiment of the invention should be understood only by the claims set forth below, all equivalent or equivalent modifications will be within the scope of the invention idea.

510: passing point recognition unit 520: image acquisition unit
530: speed calculating unit 540: encoding unit
550: vehicle number extraction unit 560: transmission unit
570: image management unit 571: decoding unit
572: output 573: database

Claims (8)

A first step of detecting passage of a first point and a second point of the vehicle passing through the preset area;
A second step of acquiring a moving video from the first point to the second point; And
A third step of calculating a vehicle speed using the number of image frames between the first point and the second point, an interframe time, and a distance between the first point and the second point;
Vehicle image detection method comprising a. The method of claim 1,
The first step of detecting,
Generating a background image every frame using a Gaussian mixture model;
Detecting a passage of the first point, which is a starting point at which the vehicle enters, by comparing the current frame with a starting point area set at the top of the image; And
Detecting the passage of the second point, which is an end point at which the vehicle disappears, by comparing the end point area set at the bottom of the image with the current frame;
Vehicle image detection method comprising a. The method of claim 1,
A fourth step of acquiring a still image for recognizing a license plate of a vehicle from a passing video from the first point to the second point;
A fifth step of recognizing the license plate in the acquired license plate recognition still image;
Generating a compressed video clip by storing the video frame corresponding to the first point and the video frame corresponding to the second point as a video clip, and encoding the stored video clip through a predetermined video compression technique; And
Transmitting traffic information including the compressed video clip, the calculated speed information of the vehicle, and the recognized license plate information to an image management unit which outputs and manages an image;
Vehicle image detection method further comprising. The method of claim 3,
In the fourth step,
The vehicle image detection method, characterized in that to obtain a still image for the middle of the moving image from the first point to the second point as a still image for the license plate recognition of the vehicle. The method of claim 3,
Decoding the received compressed video clip and outputting traffic information including the vehicle number and speed from the image manager; And
9 steps of storing the decoded video and traffic information in a database;
Vehicle image detection method further comprising. A passing point recognizing unit detecting a passing of the first point and the second point of the vehicle passing through the preset area;
An image acquisition unit for obtaining a passing video from the first point to the second point;
A speed calculator configured to calculate a vehicle speed by using the number of image frames between the first point and the second point, an interframe time, and a distance between the first point and the second point; And
Image management unit for outputting and storing at least one of the obtained video and the speed of the vehicle
Vehicle image detection system comprising a. The method according to claim 6,
A vehicle number extracting unit for obtaining a still image for license plate recognition of the vehicle from the passing video from the first point to the second point and recognizing the license plate in the acquired license plate recognition still image;
An encoding unit which stores the image frame corresponding to the first point and the image frame corresponding to the second point as a video clip, and generates a compressed video clip by encoding the stored video clip through a predetermined video compression technique; And
Transmission unit for transmitting the compressed video clip and the recognized license plate information to the image management unit
Vehicle image detection system further comprising. The method of claim 7, wherein
The image management unit,
A decoding unit for decoding the received image;
An output unit for outputting a screen of traffic information including the vehicle number and speed; And
A database storing the decoded image and the traffic information
Vehicle image detection system comprising a.

Images