KR20120114099A - Apparatus and method for car type recognition and method for coming and going managing method based on car type recognition - Google Patents

Abstract

PURPOSE: A vehicle sensing device, a method thereof and a vehicle access managing method based on vehicle recognition are provided to accurately distinguish a vehicle by recognizing the vehicle based on extracted vehicle feature information. CONSTITUTION: A camera(10) takes a photograph of a vehicle. A controller(11) extracts vehicle feature information and recognizes the vehicle. A storage(15) stores reference vehicle contour information and reference vehicle feature information. A first vehicle sensing unit(110) determines the model of the vehicle. A second vehicle sensing unit(112) compares the extracted vehicle feature information and the reference vehicle feature information. The second vehicle sensing unit recognizes the vehicle. [Reference numerals] (10) Camera; (110) First vehicle sensing unit; (112) Second vehicle sensing unit; (12) Vehicle detecting sensor; (13) Operation unit; (14) Output unit; (15) Storage; (16) Lighting unit; (17) Communication unit

Description

Apparatus and method for car type recognition and method for coming and going managing method based on car type recognition}

One aspect of the present invention relates to a vehicle recognition technology, and more particularly, to a vehicle recognition technology using a camera.

As social structures become more complex and transportation develops, various types of vehicles operate on roads and parking lots. Each vehicle is classified according to the use, such as a passenger car or a van, and may also be classified into a light car, a small car, a medium-sized car, etc. according to the displacement.

Meanwhile, as environmental pollution by vehicles increases, road running and parking fees vary according to vehicle types in order to reduce possible environmental pollution and to efficiently use road resources. Therefore, if the vehicle can be automatically recognized, it will be very convenient to collect road driving fare or parking management fee.

When a vehicle is detected, a technology for automatically recognizing a vehicle model has been proposed by photographing the front surface of the vehicle's license plate. However, the above-described technique has a limitation in accurately recognizing a vehicle. The same model may not recognize the license plate of the vehicle, or may not recognize the license plate of the vehicle properly depending on the characteristics of the country, such as the United States. For example, in the United States, the license plate configuration may differ depending on the state, and the license plate may be configured or changed in various patterns or shapes depending on whether additional costs are paid or the user's preference. Accordingly, as the number of entrances and exits of the vehicle increases, the vehicle unrecognition rate or the misrecognition rate may increase as the complete unmanned system.

According to an aspect, a vehicle recognition technology and a vehicle recognition based vehicle access management technology capable of accurately recognizing a vehicle are proposed.

According to an aspect, an apparatus for recognizing a vehicle may include extracting and extracting vehicle feature information including at least one of vehicle outline information, glass feature information, bumper feature information, and defect information from a camera photographing a vehicle and image data captured by the camera. And a controller configured to recognize the vehicle based on the vehicle feature information.

Meanwhile, a method of recognizing a vehicle by a vehicle recognizing apparatus according to another aspect includes photographing a vehicle with a camera, extracting a vehicle outline from image data captured by the camera, and analyzing the extracted vehicle outline to determine a vehicle type of the vehicle. And extracting vehicle feature information including at least one of glass feature information, bumper feature information, and defect information of the vehicle from the image data photographed by the camera, and analyzing the extracted vehicle feature information to recognize the vehicle. .

Meanwhile, a method of managing vehicle access based on vehicle recognition of a vehicle recognition apparatus according to another aspect includes detecting a vehicle entering a preset area using a vehicle detection sensor, and when the vehicle is detected, a camera of the vehicle Photographing the vehicle, extracting the vehicle outline from the image data captured by the camera, and analyzing the extracted vehicle outline to determine the vehicle model, and glass feature information and bumper feature information of the vehicle from the image data captured by the camera. And recognizing the vehicle by extracting the vehicle feature information including at least one of defect information and analyzing the extracted vehicle feature information, allowing the vehicle to enter according to the recognition result, and simultaneously identifying the vehicle type, the vehicle recognition information, and the access time. And storing vehicle information including a.

On the other hand, the vehicle access management method according to an additional aspect, the step of detecting a vehicle leaving the predetermined area using a vehicle sensor, if the vehicle is detected by taking a camera of the camera, and the image data captured by the camera Extracting the vehicle outline from the vehicle and analyzing the extracted vehicle outline to determine the vehicle model, and vehicle feature information including at least one of glass feature information, bumper feature information, and defect information of the vehicle from image data captured by the camera; Extracting and analyzing the extracted vehicle feature information and recognizing the vehicle, comparing the recognition result with the stored vehicle information, and if the comparison is the same as the vehicle type and the vehicle recognition information, permits the vehicle to be released and is charged. It includes a step.

According to an embodiment, after photographing a vehicle entering or driving, vehicle feature information including at least one of vehicle outline information, glass feature information, bumper feature information, and defect information is extracted from the captured image data, and the extracted vehicle is extracted. As the vehicle is recognized based on the feature information, the vehicle can be accurately identified.

Furthermore, the first vehicle recognition process for discriminating the vehicle type from the image data captured by the camera, and the vehicle feature information is extracted from the captured image data when the vehicle model is determined, and then combined with the vehicle type determination result to finalize the vehicle recognition. By performing the secondary vehicle recognition process to confirm, the vehicle can be correctly identified without errors.

Furthermore, the present invention can be usefully applied to a parking management system, a vehicle inspection system, a toll collection system, and the like through vehicle recognition, and in particular, when applied to the parking management system, a person can support an unmanned processing system without directly issuing a parking ticket. have.

1 is a block diagram showing a vehicle recognition apparatus according to an embodiment of the present invention,
2 is a detailed configuration diagram of the second vehicle recognition unit of FIG. 1 according to an embodiment of the present disclosure;
3 is a flowchart illustrating a vehicle recognition method of a vehicle recognition apparatus according to an embodiment of the present invention;
4 is a flowchart illustrating a vehicle model discrimination process of a first vehicle recognition unit according to an embodiment of the present invention;
5 is a flowchart illustrating a method of recognizing a vehicle based on glass feature information according to an embodiment of the present invention;
6 is a flowchart illustrating a bumper feature-based vehicle recognition process according to an embodiment of the present invention;
7 is a flowchart illustrating a driver recognition information based vehicle recognition process according to an embodiment of the present invention;
8 is a flowchart illustrating a license plate based vehicle recognition process according to an embodiment of the present invention;
9 is a configuration diagram of an example in which the present invention is applied to a parking management system;
10 and 11 are flowcharts illustrating an example process of applying the present invention to a parking management system.

Hereinafter, with reference to the accompanying drawings will be described embodiments of the present invention; In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear. In addition, the terms described below are defined in consideration of the functions of the present invention, and this may vary depending on the intention of the user, the operator, or the like. Therefore, the definition should be based on the contents throughout this specification.

1 is a block diagram showing a vehicle recognition device 1 according to an embodiment of the present invention.

Referring to FIG. 1, the vehicle recognition apparatus 1 includes a controller 11 including a camera 10, a first vehicle recognition unit 110, and a second vehicle recognition unit 112, and a vehicle detection sensor ( 12), the operation unit 13, the output unit 14, the storage unit 15, the lighting unit 16 and the communication unit 17 may be further included.

The vehicle recognition apparatus 1 of the present invention may be for recognizing a vehicle for a vehicle entering or driving in a parking management system, a toll collection system, or a vehicle inspection system. It is not limited.

The vehicle sensor 12 of the vehicle recognition device 1 detects a vehicle entering or driving. The vehicle sensor 12 may be a sensor for detecting a geomagnetic signal, an infrared signal, an ultrasonic signal, a radar signal, a laser signal, an RF signal, a temperature detection signal, a vibration detection signal, or the like of the vehicle. However, this is only one embodiment, the vehicle sensor 12 may use any vehicle detection scheme that exists now or in the future in the future. The vehicle detection sensor 12 transmits a detection signal to the controller 11 when a vehicle is detected.

The camera 10 photographs the vehicle when the vehicle is detected by the vehicle sensor 12. The camera 10 includes a lens and a CCD or CMOS image sensor, and photographs the vehicle when the vehicle is detected under the control of the controller 11. It is preferable that the camera 10 is capable of controlling a photographing posture or a photographing position by a motor or mechanical components not shown. Furthermore, the lighting unit 16 may be used to provide an optimal exposure when taking a camera. The lighting unit 16 provides illumination to facilitate photographing of the camera 10. The communication unit 17 provides an interface for communicating with an external host or network.

The control unit 11 controls the operation of each element of the vehicle recognition device 1. That is, the control unit 11 is connected to an external host or network through the communication unit 17 to update the internal data including the reference vehicle contour information for each vehicle type and the reference vehicle characteristic information for each vehicle accessory stored in the storage unit 15. The reference vehicle contour information is feature point information that is characterized by vehicle types. The reference vehicle feature information for each vehicle part is feature point information that is a unique feature for each vehicle part, for example, glass or bumper.

When the vehicle is detected by the vehicle sensor 12, the controller 11 controls the camera 10 to photograph the vehicle, and the first vehicle recognition unit 110 and the second vehicle recognition unit 112 from the captured image. Recognize the vehicle through). In addition, various initial values are set according to the user control input through the operation unit 13, an operation mode is set, and an image, an operation state, or a vehicle recognition result photographed is displayed on the output unit 14.

The controller 11 extracts vehicle contour information, vehicle feature information including glass feature information, bumper feature information, and defect information from the image data photographed by the camera 10, and based on the extracted vehicle feature information. Recognize.

In more detail, the controller 11 includes a first vehicle recognition unit 110 and a second vehicle recognition unit 112. The first vehicle recognition unit 110 extracts the vehicle outline from the image data photographed by the camera 10. The vehicle model of the vehicle is determined by comparing and analyzing the extracted vehicle contour with reference vehicle contour information stored in the storage unit 15. According to an exemplary embodiment, the first vehicle recognition unit 110 receives the image data captured by the camera 10 to process filtering, background removal, contour extraction, and image matching (compare) to recognize a vehicle model of the vehicle. A vehicle model recognition process of the vehicle of the first vehicle recognition unit 110 will be described later with reference to FIG. 4.

The second vehicle recognition unit 112 extracts vehicle feature information from the image data photographed by the camera 10, and compares the extracted vehicle feature information with reference vehicle feature information stored in the storage unit 15 to recognize the vehicle. do. In this case, the vehicle feature information includes glass feature information, bumper feature information, and defect information of the vehicle.

According to an exemplary embodiment, when the vehicle type of the vehicle is determined through the first vehicle recognition unit 110, the second vehicle recognition unit 112 may determine the vehicle feature information from the image data photographed by the camera 10 with respect to the determined vehicle. Extract In addition, the vehicle feature information may be combined with the vehicle type determination result through the first vehicle recognition unit 110 and analyzed to determine the vehicle recognition.

The storage unit 15 stores reference vehicle contour information for each vehicle type and reference vehicle feature information for each vehicle part. The reference vehicle contour information by vehicle type and the reference vehicle feature information by vehicle parts may be periodically updated or updated through the communication unit 17 whenever a new vehicle or accessory is released.

2 is a detailed block diagram of the second vehicle recognizing unit 112 of FIG. 1 according to an exemplary embodiment of the present invention.

The second vehicle recognizer 112 includes a glass feature information based vehicle recognizer 1120, a bumper feature information based vehicle recognizer 1122, a driver feature information based vehicle recognizer 1124, and a license plate based vehicle recognizer 1126. It includes.

The glass feature information-based vehicle recognition unit 1120 extracts vehicle glass feature information from the image data captured by the camera 10 to recognize the vehicle. The vehicle glass feature information includes at least one of the location, type or size of the attachment attached to the glass and the location, type or size of the defects in the glass. The vehicle recognition process of the glass feature information-based vehicle recognition unit 1120 will be described later with reference to FIG. 5.

The bumper feature information-based vehicle recognition unit 1122 recognizes the vehicle by extracting the vehicle bumper feature information from the image data photographed by the camera 10. And at least one of the location, type, or size of the attachment attached to the bumper, and the location, type, or size of the defect location of the bumper. The vehicle recognition process of the bumper feature information-based vehicle recognition unit 1122 will be described later with reference to FIG. 6.

The driver characteristic information-based vehicle recognition unit 1124 extracts driver characteristic information from the image data photographed by the camera 10 to recognize the vehicle. The vehicle recognition process of the driver feature information based vehicle recognition unit 1124 will be described later with reference to FIG. 7.

The license plate-based vehicle recognition unit 1126 recognizes the vehicle by extracting the license plate character from the image data of the vehicle. The vehicle recognition process of the license plate based vehicle recognition unit 1126 will be described later with reference to FIG. 8.

According to an embodiment, the second vehicle recognition unit 112 may extract the glass feature information extracted from the glass feature information-based vehicle recognition unit 1120 based on the vehicle model information recognized by the first vehicle recognition unit 110. And bumper feature information extracted from the bumper feature information-based vehicle recognizing unit 1122, driver feature information extracted from the driver feature information-based vehicle recognizing unit 1124 and license plate extracted from the license plate-based vehicle recognizing unit 1126. Character recognition information may be analyzed individually or in combination to finally determine vehicle recognition.

For example, the second vehicle recognizing unit 112 temporarily stores the vehicle model information determined by the first vehicle recognizing unit 110 as the primary key value in the storage unit 15. Subsequently, the glass feature information extracted from the glass feature information based vehicle recognition unit 1120 and the driver feature information extracted from the driver feature information based vehicle recognition unit 1124 are temporarily stored in the storage unit 15 as a secondary key value. . The bumper feature information based on the vehicle recognition unit 1122 is temporarily stored as a third key value. Then, the primary, secondary and tertiary key values are combined and stored in the storage unit 15 as vehicle identification feature values, and the vehicle is identified from the stored vehicle identification feature values. In this case, the license plate character recognition information extracted from the license plate-based vehicle recognition unit 1126 may be combined with the aforementioned vehicle identification feature to finally identify the vehicle. Meanwhile, the above-described example is only one embodiment for better understanding of the present invention, and the vehicle recognition embodiment through individual or combination analysis may be variously implemented.

3 is a flowchart illustrating a vehicle recognition method of the vehicle recognition apparatus 1 according to an embodiment of the present invention.

Referring to FIG. 3, the vehicle recognition apparatus 1 photographs a vehicle with the camera 10 (300). For example, the front surface of the vehicle is photographed by the camera 10. In recent years, the camera technology has been advanced, so that the photographed digital data can be directly provided in a raw file format or various standard files (jpg, etc.). Subsequently, the vehicle outline is extracted 310 from the image data captured by the camera 10, and the extracted vehicle outline is analyzed to determine a vehicle model of the vehicle (320). Subsequently, vehicle feature information including glass feature information, bumper feature information, and defect information of the vehicle is extracted from the image data photographed by the camera (330), and the extracted vehicle feature information is analyzed to recognize the vehicle (340).

4 is a flowchart illustrating a vehicle model determining process of the first vehicle recognizing unit 110 according to an exemplary embodiment of the present invention.

Referring to FIG. 4, the first vehicle recognition unit 110 receives data photographed by the camera 10, performs preprocessing or filtering, removes a background, and extracts outlines (400 to 420). Subsequently, the vehicle outline and the reference contour information stored in the storage unit 15 are compared and matched with each other (430) to determine the vehicle as the vehicle model of the reference outline that is the most close (440).

5 is a flowchart illustrating a method of recognizing a vehicle based on glass feature information according to an embodiment of the present invention.

Referring to FIG. 5, the glass feature information-based vehicle recognition unit 1120 extracts a glass region from image data captured by the camera 10 (500).

Subsequently, the vehicle glass feature information is extracted by comparing the extracted glass area with reference glass feature information stored in the storage 15 (510) (520). Reference glass feature information includes features such as the type, type, or size of glass that is standard for each vehicle type or glass accessory. The extracted vehicle glass feature information may be the position, type or size of the attachment attached to the glass, and the defect location, type or size of the glass. For example, the attachment may be various stickers or driver personal accessories attached to the glass front. For example, it may be a sticker that can be identified as 'disabled person' attached to the front of the glass of the vehicle.

Subsequently, the vehicle feature-based vehicle recognition unit 1120 recognizes the vehicle using the extracted vehicle glass feature information (530). For example, when a mark on the front surface of the vehicle is marked with a mark for identifying that the sticker is for the handicapped only, the glass feature information-based vehicle recognition unit 1120 identifies the vehicle from the mark for the handicapped only. can do.

6 is a flowchart illustrating a bumper feature-based vehicle recognition process according to an embodiment of the present invention.

Referring to FIG. 6, the bumper feature information-based vehicle recognizer 1122 extracts a bumper region from image data captured by the camera 10 (600).

Subsequently, the vehicle bumper feature information is extracted (620) by comparing the extracted bumper area with reference bumper feature information (610). The reference bumper feature information includes features on the type, type, or size of a bumper, which is a standard for each vehicle type or bumper accessory. The extracted vehicle bumper feature information includes the position, type or size of the attachment attached to the bumper, and the defect position, type or size of the bumper. For example, if a specific portion of the vehicle bumper is crushed, the position and size of the crushed portion may be extracted as the vehicle bumper feature information.

Subsequently, the vehicle recognition unit 1122 based on the bumper feature information recognizes the vehicle by using the extracted vehicle bumper feature information (630). For example, when a certain portion of the bumper is crushed, the bumper characteristic information-based vehicle recognition unit 1122 may identify the vehicle from the crushed shape of the bumper.

7 is a flowchart illustrating a driver recognition information based vehicle recognition process according to an embodiment of the present invention.

Referring to FIG. 7, the driver characteristic information-based vehicle recognition unit 1124 recognizes a face through a predetermined face recognition algorithm from the image data photographed by the camera 10 and extracts driver characteristic information (700, 710). The face recognition algorithm extracts feature points through Harrison, Hessian corner detection, or DoG method, and obtains feature information of extracted feature points through methods such as generalized color moments or SIFT. It is possible. For example, the driver's characteristic information may be, but is not limited to, a driver's hairstyle, glasses or sunglasses, or a fashion style.

Subsequently, the driver characteristic information-based vehicle recognition unit 1124 recognizes the vehicle by analyzing the extracted driver characteristic information and the vehicle characteristic information including the glass characteristic information, the bumper characteristic information, and the defect information individually or in combination (720). .

8 is a flowchart illustrating a license plate based vehicle recognition process according to an embodiment of the present invention.

Referring to FIG. 8, the license plate-based vehicle recognition unit 1126 extracts a license plate area from image data of a vehicle (800). After extracting the text area from the extracted license plate area (810), the license plate character is recognized according to a predetermined character recognition algorithm (820). Subsequently, the vehicle is recognized by analyzing the license plate character information, the vehicle feature information including the glass feature information, the bumper feature information, and the defect information, individually or in combination.

9 is a configuration diagram of an example in which the present invention is applied to a parking management system.

Referring to FIG. 9, the parking management system includes a parking controller 910 installed at an entrance of a parking lot, a vehicle sensor 911, a camera 912, a breaker 914, a parking controller 930 installed at an exit of a parking lot, The camera 931, the charge display unit 932, the breaker 933, the vehicle sensor 934, and the parking management server 920 installed in the management room are included.

10 and 11 are flowcharts illustrating an example process of applying the present invention to a parking management system. In this case, for the convenience of description, reference numerals described above are referred to in FIG. 9.

Referring to FIG. 10, at a parking lot entrance, the vehicle detection sensor 911 detects a vehicle entering the parking lot and transmits the detected vehicle 1000 to the entrance controller 910, and the entrance controller 910 receives the vehicle from the vehicle detection sensor 911. If detected, the camera 912 is controlled to photograph the entry vehicle 1010.

Subsequently, the entrance controller 910 extracts a vehicle outline from the image data captured by the camera 912 (1020) and analyzes the extracted vehicle outline to determine a vehicle type of the vehicle (1030).

Subsequently, the parking controller 910 extracts 1040 vehicle feature information including glass feature information, bumper feature information, and defect information of the vehicle from the image data photographed by the camera 912, and analyzes the extracted vehicle feature information. The vehicle is recognized (1050).

Subsequently, the entrance controller 910 controls the breaker 914 according to the recognition result to allow entry of the vehicle (1060). In operation 1060, vehicle information including the vehicle model, vehicle identification information, and access time is stored in the parking management server 920.

Referring to FIG. 11, at the exit of the parking lot, the vehicle sensor 934 detects a vehicle exiting from the parking lot 1100, and delivers the vehicle to the departure controller 930, and the departure controller 930 receives the vehicle from the vehicle detection sensor 934. When the detection is detected, the camera 931 is controlled to photograph the exiting vehicle (1110).

Subsequently, the departure controller 930 extracts the vehicle outline from the image data photographed by the camera 934 (1120) and analyzes the extracted vehicle outline to determine the vehicle model of the vehicle (1130).

Subsequently, the departure controller 930 extracts the vehicle characteristic information including glass characteristic information, bumper characteristic information, and defect information of the vehicle from the image data photographed by the camera 934 (1140), and analyzes the extracted vehicle characteristic information. The vehicle is recognized (1150).

Subsequently, the parking controller 930 obtains the parking information of the vehicle from the parking management server 920 and compares it with the vehicle recognition result (1160). As a result of comparison, when the vehicle model and the vehicle identification information coincide with each other, the circuit breaker 933 is controlled to allow the vehicle to be taken out and the charging is performed (1170).

According to one embodiment, the parking controller 930 calculates the parking fee until the departure time, displays the parking fee on the charge display unit 932 and processes the settlement. At this time, in the case of a light car using the recognized car model information, a parking discount may be calculated by applying a predetermined discount. When the settlement is completed, the breaker 933 is opened to allow the vehicle to exit, and the settlement related information is transmitted to the parking management server 920.

In the example of the parking management system to which the present invention is applied, vehicle type recognition is described as being performed in all the entering and exiting stages, but in addition, vehicle type recognition may be performed only at the entering stage, and vehicle model recognition may be processed only at the leaving stage.

The embodiments of the present invention have been described above. Those skilled in the art will appreciate that the present invention can be implemented in a modified form without departing from the essential features of the present invention. Therefore, the disclosed embodiments should be considered in an illustrative rather than a restrictive sense. The scope of the present invention is defined by the appended claims rather than by the foregoing description, and all differences within the scope of equivalents thereof should be construed as being included in the present invention.

1: vehicle recognition device 10: camera
11 control unit 12 vehicle detection sensor
13: operation unit 14: output unit
15: storage unit 16: lighting unit
17: communication unit 110: first vehicle recognition unit
112: second vehicle recognition unit
1020: Vehicle feature based on glass feature information
1122: bumper feature information based vehicle recognition unit
1124: vehicle recognition unit based on driver characteristic information
1126: license plate based vehicle recognition unit

Claims (18)

A camera for photographing the vehicle; And
Extracting vehicle feature information including at least one of vehicle outline information, glass feature information, bumper feature information, and defect information from the image data photographed by the camera, and recognizing the vehicle based on the extracted vehicle feature information. Control unit;
Vehicle identification apparatus comprising a. The method of claim 1,
Further comprising a storage unit for storing the reference vehicle contour information by vehicle type and the reference vehicle feature information by vehicle parts,
The control unit,
A first vehicle recognition unit extracting a vehicle outline from the image data photographed by the camera and comparing and analyzing the extracted vehicle outline with reference vehicle outline information stored in the storage unit to determine a vehicle type of the vehicle; And
And when the vehicle model of the vehicle is determined, extracting the vehicle feature information from the image data photographed by the camera, and comparing the extracted vehicle feature information with reference vehicle feature information stored in the storage unit to recognize the vehicle. 2 vehicle recognition unit;
Vehicle identification apparatus comprising a. The method of claim 2, wherein the second vehicle recognition unit,
Extracting a glass region from the image data photographed by the camera, and comparing the extracted glass region with reference glass characteristic information among the reference vehicle characteristic information stored in the storage unit; A glass feature information-based vehicle recognition unit extracting vehicle glass feature information including at least one of a defect position, type, or size of glass, and recognizing the vehicle using the extracted vehicle glass feature information;
Vehicle identification apparatus comprising a. The method of claim 2, wherein the second vehicle recognition unit,
Extracting the bumper area from the image data photographed by the camera, comparing the extracted bumper area with the reference bumper feature information of the reference vehicle feature information, the position, type or size of the attachment attached to the bumper and the defect position of the bumper A bumper feature information-based vehicle recognition unit extracting vehicle bumper feature information including at least one of a type and a size, and recognizing the vehicle using the extracted vehicle bumper feature information;
Vehicle identification apparatus comprising a. The method of claim 2, wherein the second vehicle recognition unit,
Driver features for extracting driver feature information from the image data photographed by the camera according to a predetermined face recognition algorithm and analyzing the extracted driver feature information and the extracted vehicle feature information individually or in combination to recognize the vehicle. An information-based vehicle recognition unit;
Vehicle identification apparatus comprising a. The method of claim 2,
Extracting the license plate area from the image data taken by the camera, extracting the character area from the extracted license plate area, recognizing the license plate character according to a predetermined character recognition algorithm, the recognized license plate character information and the extracted vehicle A license plate-based vehicle recognition unit for recognizing the vehicle by analyzing feature information individually or in combination;
Recognizing apparatus of a vehicle comprising a. The method of claim 2, wherein the second vehicle recognition unit
A driver feature information based vehicle recognition unit configured to extract driver feature information from the image data according to a predetermined face recognition algorithm;
And extracting the license plate area from the image data, extracting the character area from the extracted license plate area, and then recognizing the license plate character according to a predetermined character recognition algorithm.
The second vehicle recognizing unit recognizes the vehicle by analyzing the glass feature information, the bumper feature information, the defect information, the driver feature information, and the license plate character recognition information of the vehicle in which the vehicle model is determined, individually or in combination. Vehicle identification device. 8. The vehicle recognition apparatus according to any one of claims 1 to 7, wherein
A vehicle recognition device, which is used for a parking management system, a toll collection system, or a vehicle inspection system. In the vehicle recognition apparatus recognizes a vehicle,
Photographing the vehicle with a camera;
Extracting a vehicle outline from the image data captured by the camera and analyzing the extracted vehicle outline to determine a vehicle type of the vehicle; And
Extracting vehicle feature information including at least one of glass feature information, bumper feature information, and defect information of the vehicle from the image data photographed by the camera, and analyzing the extracted vehicle feature information to recognize the vehicle;
Vehicle recognition method comprising a. The method of claim 9, wherein the analyzing of the extracted vehicle characteristic information to recognize the vehicle comprises:
When the vehicle type of the vehicle is determined, the vehicle feature information is extracted from the image data photographed by the camera, and the extracted vehicle feature information is analyzed individually or in combination with the vehicle type determination result. Vehicle identification method, characterized in that the final determination of the vehicle recognition. The method of claim 9, wherein the analyzing of the extracted vehicle characteristic information to recognize the vehicle comprises:
Extracting a glass area from the image data captured by the camera;
Comparing the extracted glass region with previously stored reference glass characteristic information to extract vehicle glass characteristic information including at least one of a position, a kind or a size of an attachment attached to the glass, and a defect position, a kind, or a size of the glass; step; And
Recognizing the vehicle using the extracted vehicle glass feature information;
Vehicle recognition method comprising a. The method of claim 9, wherein the analyzing of the extracted vehicle characteristic information to recognize the vehicle comprises:
Extracting a bumper area from the image data photographed by the camera;
Comparing the extracted bumper area with previously stored reference bumper feature information to extract vehicle bumper feature information including at least one of a position, a type or a size of an attachment attached to the bumper, and a defect location, type or size of the bumper; ; And
Recognizing the vehicle using the extracted vehicle bumper feature information;
Vehicle recognition method comprising a. The method of claim 9,
Extracting driver characteristic information from the image data according to a predetermined face recognition algorithm,
Recognizing the vehicle by analyzing the extracted vehicle feature information,
And recognizing the vehicle by analyzing the extracted driver characteristic information and the extracted vehicle characteristic information of the vehicle in which the vehicle model of the vehicle is determined, individually or in combination. The method of claim 9,
Extracting a license plate area from the image data; And
Extracting a character area from the extracted license plate area and recognizing the license plate letter according to a predetermined character recognition algorithm;
Recognizing the vehicle by analyzing the extracted vehicle feature information,
And recognizing the vehicle by analyzing the extracted vehicle characteristic information and the license plate character recognition information of the vehicle in which the vehicle model of the vehicle is determined, individually or in combination. The method of claim 9,
Extracting driver characteristic information from the image data according to a predetermined face recognition algorithm;
Extracting a license plate region from the image data, extracting a character region from the extracted license plate region, and recognizing the license plate character according to a predetermined character recognition algorithm,
Recognizing the vehicle by analyzing the extracted vehicle feature information,
And recognizing the vehicle individually or in combination by analyzing glass feature information, bumper feature information, defect information, driver feature information, and license plate character recognition information of the vehicle in which the vehicle model of the vehicle is determined. In the method for managing vehicle access based on the vehicle recognition of the vehicle recognition device,
Detecting a vehicle entering a preset area using a vehicle detection sensor;
Photographing the vehicle with a camera when the vehicle is detected;
Extracting a vehicle outline from the image data captured by the camera and analyzing the extracted vehicle outline to determine a vehicle type of the vehicle;
Extracting vehicle feature information including at least one of glass feature information, bumper feature information, and defect information of the vehicle from the image data photographed by the camera, and analyzing the extracted vehicle feature information to recognize the vehicle; And
Allowing vehicle access according to the recognition result and storing vehicle information including vehicle type, vehicle identification information, and access time;
Vehicle recognition based vehicle access management method comprising a. 17. The method of claim 16,
Detecting a vehicle leaving the preset area using the vehicle detection sensor;
Photographing the vehicle with a camera when the vehicle is detected;
Extracting a vehicle outline from the image data captured by the camera and analyzing the extracted vehicle outline to determine a vehicle type of the vehicle;
Extracting vehicle feature information including at least one of glass feature information, bumper feature information, and defect information of the vehicle from the image data photographed by the camera, and analyzing the extracted vehicle feature information to recognize the vehicle;
Comparing the recognition result with the stored vehicle information; And
Allowing the vehicle to leave the vehicle and granting the charge if the vehicle type and the vehicle identification information match each other as a result of the comparison;
Vehicle recognition based vehicle access management method further comprising. The method of claim 17,
Extracting driver characteristic information from the image data according to a predetermined face recognition algorithm; And
Extracting the license plate area from the image data, extracting the character area from the extracted license plate area, and recognizing the license plate character according to a predetermined character recognition algorithm,
Recognizing the vehicle by analyzing the extracted vehicle feature information,
Vehicle recognition-based vehicle access characterized in that the vehicle is recognized by analyzing the glass feature information, bumper feature information, defect information, driver feature information and license plate character recognition information individually or in combination, the vehicle type is determined How to manage.

Images