KR20140004413A - Apparatus and method for assisting safe driving - Google Patents

Abstract

The present invention relates to a device and a method for assisting safe driving to detect obstacles using the camera and rider of a vehicle for assisting drivers in driving the vehicle by determining if the detected obstacles exist on an inside lane and if the obstacles are pedestrians or vehicles; and to recognize a sign using the camera and rider of the vehicle for assisting the driver in driving the vehicle by determining the type of the sign. The device for assisting safe driving may comprise a camera information receiving part for receiving information collected by the camera of a vehicle; a rider information receiving part for receiving information collected by the rider; and an information recognition part for recognizing information relating to objects included in the information received in the camera and rider information receiving parts. [Reference numerals] (1000) Device for assisting safe driving; (1100) Camera information receiving part; (1200) Rider information receiving part; (1300) Image signal processing part; (1400) 3D pre-processing part; (1500) 2D/3D matching part; (1600) Information recognition part; (1700) Warning part

Description

Apparatus and Method for Assisting Safe Driving}

The present invention relates to a safe driving support apparatus and method. More specifically, the present invention recognizes an obstacle using a camera and a rider of the vehicle, and assists the vehicle driving by determining whether the recognized obstacle exists in the driving lane or whether the obstacle is a vehicle or a pedestrian, and the camera of the vehicle. And a safe driving support apparatus and a method for recognizing a sign using a rider and determining a type of the sign to assist in driving a vehicle.

Conventional riders (Lidar) using the vehicle, using the rider to obtain the distance information with the other vehicle, using the obtained distance information and the image received from the camera of the vehicle is different from the presence of other vehicles There is a system that recognizes the distance to the vehicle to prevent the collision between the vehicles.

However, a system for preventing collision between vehicles using such a rider does not have a function of recognizing pedestrians, and thus, when the recognized obstacle is a pedestrian instead of a vehicle, there is a problem in that an accident with a pedestrian cannot be prepared.

In addition, an obstacle recognition system using a conventional rider and a collision avoidance system using a vehicle do not distinguish a vehicle existing in a driving lane from a vehicle existing in a lane other than a driving lane.

The vehicle may have a function of acquiring an external image of the vehicle with a camera installed in the vehicle and detecting a traffic sign from the acquired image.

However, when detecting a traffic sign using only an image acquired by using a camera, there is a problem in that a false recognition rate of the traffic sign is high according to climatic conditions, day and night time, and the like.

In order to solve the above problem, the present invention recognizes a driving lane of a vehicle to determine whether an obstacle exists in the driving lane, and determines whether the recognized obstacle is a vehicle or a pedestrian by using a database to assist vehicle operation. It is an object of the present invention to provide a safe driving support apparatus and method for more accurately recognizing a sign using a rider as well as a camera to recognize a sign.

In order to solve the above problems, the safe driving support apparatus includes: a camera information receiver configured to receive information collected by a camera of a vehicle; A rider information receiver configured to receive information collected by a rider; And an information recognizing unit recognizing information on an object included in the received information by using the information received by the camera information receiving unit and the rider information receiving unit.

Preferably, the information recognizing unit includes an obstacle classification recognition unit for recognizing information on an object which is an obstacle to the running of the vehicle among the objects included in the received information, the obstacle classification recognition unit, the camera information receiving unit A lane recognizing unit recognizing a lane including a lane in which the vehicle is driven by using the received information; An obstacle recognition unit that recognizes the obstacle by using the information received from the camera information receiver and the rider information receiver; And a driving lane obstacle determining unit determining whether the recognized obstacle is an obstacle present in the driving lane recognized by the lane recognition unit.

Preferably, the obstacle classification unit may further include an obstacle division unit that determines that the recognized obstacle is any one of a pedestrian, a vehicle, and a noise.

Preferably, the safe driving support apparatus generates intensity information and depth information by using the information received from the rider information receiving unit, and generates a 3D image by using the generated information. 3D pre-processing unit, the obstacle classification unit, a first database unit for storing a plurality of pedestrian image and vehicle image, calculates a gradient (Gradient) using the camera information and the generated three-dimensional image, Acquiring edge information from the image using the calculated gradient, acquiring outline information of the recognized obstacle using the edge information, and matching the outline information with an image stored in the first database unit, thereby recognizing the edge information. The obstacle may be determined to be one of a pedestrian, a vehicle, and a noise.

Preferably, the safe driving support apparatus, when the determination of the driving lane obstacle determining unit determines that the recognized obstacle is an obstacle existing between the recognized driving lanes, a sound, a screen display, a seat belt vibration of the vehicle, or the like. The apparatus may further include an attention warning unit configured to perform at least one of brake control of the vehicle.

Preferably, the information recognizing unit includes a sign recognizing unit recognizing information on the sign of the object included in the received information, the sign recognition unit, the information received from the camera information receiving unit and the rider information receiving unit A sign candidate selection unit for selecting an image assumed to be a sign using the sign candidate; A second database unit storing a plurality of sign images; And a sign determination unit determining a sign type of the selected image by matching the image selected by the sign candidate selection unit with the sign image stored in the second database unit.

Preferably, the safe driving support apparatus generates intensity information and depth information by using the information received from the rider information receiving unit, and generates a 3D image by using the generated information. And a 3D pre-processing unit, wherein the sign candidate selection unit may select a portion of the generated 3D image having the intensity greater than or equal to a predetermined intensity or a portion of the 3D image as an image estimated to be a sign.

Preferably, the safe driving support apparatus generates intensity information and depth information by using the information received from the rider information receiving unit, and generates a 3D image by using the generated information. And a 3D preprocessor, wherein the sign determination unit extracts a line edge of the 3D image to determine a part that is assumed to be a sign part, and matches the determined sign part with the sign image stored in the second database part. The type of sign of the selected image may be determined.

Safe driving support method for solving the above problems, Camera information receiving step of receiving information collected by the camera of the vehicle; A rider information receiving step of receiving information collected by a rider; And an information recognition step of recognizing information about an object included in the received information by using the information received in the camera information receiving step and the rider information receiving step.

Preferably, the information recognizing step includes an obstacle classification recognition step of recognizing information on an object which is an obstacle to driving of the vehicle among the objects included in the received information, the obstacle classification recognition step, the camera A lane recognition step of recognizing a lane including a lane in which the vehicle is driven by using the information received in the information receiving step; An obstacle recognition step of recognizing an obstacle using information received in the camera information receiving step and the rider information receiving step; And a driving lane obstacle determining step of determining whether the recognized obstacle is an obstacle present in the traveling lane recognized in the lane recognition step.

Preferably, the obstacle classification recognition step may further include an obstacle classification step of determining that the recognized obstacle is any one of a pedestrian, a vehicle, and a noise.

Preferably, in the safe driving support method, intensity information and depth information are generated using the information received in the rider information receiving step, and a 3D image is generated using the generated information. And a 3D preprocessing step, wherein the obstacle classification step includes calculating a gradient using the camera information and the generated 3D image, obtaining edge information from the image using the calculated gradient, After acquiring the contour information of the recognized obstacle using the edge information, matching the contour information with an image stored in a first database unit storing a plurality of pedestrian images and vehicle images, the recognized obstacle is a pedestrian, It may be determined that the vehicle is either one of noise or noise.

Preferably, in the safe driving support method, when the determination of the driving lane obstacle determining step determines that the recognized obstacle is an obstacle existing between the recognized driving lane, a sound, a screen display, and a seat belt vibration of the vehicle Or a warning warning step of performing at least one of brake control of the vehicle.

Preferably, the information recognition step includes a sign recognition step of recognizing information on the sign of the objects included in the received information, the sign recognition step, the camera information receiving step and the rider information receiving step A sign candidate selection step of selecting an image assumed to be a sign using the received information; And a sign determination step of determining a sign type of the selected image by matching the image selected in the sign candidate selection step with an image stored in a second database unit in which a plurality of sign images are stored.

Preferably, in the safe driving support method, intensity information and depth information are generated using the information received in the rider information receiving step, and a 3D image is generated using the generated information. And a 3D preprocessing step, wherein the sign candidate selection step may select the 3D image or a part of the 3D image having an intensity value greater than or equal to a preset intensity among the generated 3D images as an image estimated to be a sign. have.

Preferably, in the safe driving support method, intensity information and depth information are generated using the information received in the rider information receiving step, and a 3D image is generated using the generated information. And a 3D preprocessing step, wherein the sign determination step extracts a line edge of the 3D image to determine a part estimated as a sign part, and the sign image stored in the determined sign part and the second database part. The type of sign of the selected image may be determined by matching with.

The present invention can distinguish the vehicle in the driving lane and the vehicle in the other lane to perform a collision avoidance function.

In addition, the present invention can recognize the pedestrian as well as the vehicle can prevent a collision with the pedestrian.

The present invention can recognize a sign with higher accuracy and grasp the type of sign.

1 is a block diagram of a safe driving support apparatus according to a preferred embodiment of the present invention.
2 is a block diagram illustrating an example of an information recognition unit of an obstacle classification recognition apparatus according to the present invention.
3 is a diagram illustrating an example of a result of applying Equation 1 when the obstacle recognition unit recognizes an obstacle using image information.
4 is a diagram illustrating an example of a method of extracting a sign part from an image by the sign determining unit of the safe driving support apparatus according to the present invention.
5 is a view for explaining an example of a sign image stored in a second database unit of the safe driving support apparatus according to the present invention.
6 is a flowchart illustrating a safe driving support method according to an embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings. In the following description and the accompanying drawings, substantially the same components are denoted by the same reference numerals, respectively, and redundant description will be omitted. 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.

It is to be understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, . On the other hand, when an element is referred to as being "directly connected" or "directly connected" to another element, it should be understood that there are no other elements in between.

In this specification, a singular form may include plural forms unless specifically stated in the phrase. It is noted that the terms "comprises" and / or "comprising" used in the specification are intended to be inclusive in a manner similar to the components, steps, operations, and / Or additions.

The obstacle classification recognition apparatus and the sign recognition apparatus according to the present invention use a rider. Preferably, a 3D Flash Lidar may be used.

1 is a block diagram of a safe driving support apparatus according to a preferred embodiment of the present invention.

Referring to FIG. 1, the safe driving support apparatus 1000 includes a camera information receiver 1100, a rider information receiver 1200, an image signal processor 1300, a 3D preprocessor 1400, and a 2D / 3D matcher 1500. , Information recognition unit 1600 and the attention warning unit 1700.

The camera information receiver 1100 receives information collected by a camera installed in a vehicle. The information collected by the camera may include image or image information.

The rider information receiver 1200 receives information collected by a rider.

The camera and the rider are preferably installed to easily collect information in front of the vehicle. For example, the camera and the rider may be installed directly behind the windshield of the vehicle and between the room mirror of the ceiling.

The camera information received by the camera receiver is transmitted to the image signal processor 1300. The camera information received by the camera receiver may include information obtained by photographing a field of view (FOV) area of the rider as a 2D image.

The image signal processor 1300 receives the camera information, changes the size and format to improve the processing speed of the image or the image included in the camera information, and performs an image quality improvement operation. The image signal processor 1300 changes the size and format and transmits the information on the picture enhancement to the 2D / 3D matching unit 1500. The information transmitted from the image signal processor 1300 to the 2D / 3D matching unit 1500 may be 2D image information.

The 2D / 3D matching unit 1500 extracts an image of the ROI using information received from the image signal processor 1300 and information received from the 3D preprocessor 1400. More detailed description will be described again after the 3D preprocessor 1400 is described.

The 3D preprocessor 1400 receives the information collected by the rider in the rider information receiver 1200 and generates intensity information using the intensity of the position and the wavelength received by the light receiving array unit of the rider. do. In addition, the 3D preprocessor 1400 generates depth information according to the time received by the light receiving array unit of the rider.

The 3D preprocessor 1400 may generate a 3D volume image using the generated intensity information and the depth information.

The 2D / 3D matching unit 1500 may extract an image of the ROI by using the 2D image information received from the image signal processor 1300 and the 3D volume image generated by the 3D preprocessor 1400.

The image extracted by the 2D / 3D matching unit 1500 may be transmitted to the information recognizing unit 1600 and used.

The information recognizer 1600 receives information from the image signal processor 1300, the 3D preprocessor 1400, or the 2D / 3D matching unit 1500, and recognizes information about an object included in the received information.

In more detail, the information recognizing unit 1600 may recognize information about an object or a sign that is an obstacle to the operation of the vehicle, from information on an object included in the received information.

2 is a block diagram illustrating an example of an information recognition unit of an obstacle classification recognition apparatus according to the present invention.

2, the information recognizing unit 1600 may include an obstacle classification recognizing unit 1610 and a sign recognizing unit 1620.

The obstacle classification unit 1610 obtains information about an object that is an obstacle to driving of the vehicle by using the information received from the image signal processor 1300, the 3D preprocessor 1400, or the 2D / 3D matching unit 1500. .

In more detail, the obstacle classification unit 1610 may include the lane recognition unit 1611, the obstacle recognition unit 1613, the driving lane obstacle determination unit 1615, the obstacle classification unit 1617, and / or the first database unit 1619. It may include.

The lane recognizing unit 1611 recognizes a lane in which the vehicle is driven by using information received from the camera information receiving unit 1100 or information processed by the image signal processing unit 1300. The lane recognizing unit 1611 may recognize a lane and may use a known lane recognition algorithm, and thus a detailed description thereof will be omitted.

The obstacle recognition unit 1613 recognizes the obstacle using information generated by the image signal processor 1300, the 3D preprocessor 1400, and the 2D / 3D matching unit 1500.

Specifically, when the obstacle recognition unit 1613 recognizes the obstacle by using the image information generated by the image signal processor 1300, the 3D preprocessor 1400, or the 2D / 3D matching unit 1500, the ROI in the image. Obstacle recognition can be performed by cutting only the image of.

The obstacle recognition unit 1613 may extract edge information of the image pixel from the image information generated by the image signal processor 1300, the 3D preprocessor 1400, or the 2D / 3D matching unit 1500. The obstacle recognition unit 1613 extracts the edge information of the image pixel using a gradient calculation formula as shown in Equation (1).

By accumulating G (x, y), which is a size value, based on angle information for each pixel, and dividing it by the total pixel value, a gradient histogram (HOG) can be obtained.

3 is a diagram illustrating an example of a result of applying Equation 1 when the obstacle recognition unit recognizes an obstacle using image information.

Referring to FIG. 3, when the obstacle recognition unit 1613 applies Equation 1 to an image, it can be seen that contour information of a specific object is displayed in the image.

That is, (b) represents the contour information obtained by applying the equation (1) in the image (a). (D) represents the contour information obtained by applying the equation (1) in the image (c).

The contour information may be compared with a plurality of pedestrian images and vehicle images stored in the first database unit 1619 to determine whether the vehicle is a pedestrian or a pedestrian.

In detail, numerous pedestrian images and numerous vehicle images may be stored in the first database unit 1619.

The obstacle classification unit 1617 may include a pedestrian recognition unit 1617a, a vehicle recognition unit 1617c, and a noise recognition unit 1617e.

The pedestrian recognition unit 1617a may compare the contour information with the image stored in the first database unit 1619 to determine whether the recognized obstacle is a pedestrian. That is, the pedestrian recognition unit 1617a may recognize the obstacle as the pedestrian when the contour information and the image and the matching rate of the image stored in the first database unit 1619 are greater than or equal to a preset value.

The vehicle recognition unit 1617c may determine whether the recognized obstacle is a vehicle by comparing the contour information with an image stored in the first database unit 1619. That is, the vehicle recognition unit 1617c may recognize the obstacle of the outline as a vehicle when the contour information and the image and the matching ratio of the image stored in the first database unit 1619 are greater than or equal to a preset value.

If there is no image in which the outline of the recognized obstacle corresponds to the image stored in the first database unit 1619 or more than the preset value, the noise recognition unit 1617e may determine that the recognized obstacle is due to noise, not pedestrian or vehicle. Judging by it. If it is determined that the noise, the attention warning unit 1700 may be unnecessary to control the driver or other vehicles, pedestrian warning.

The driving lane obstacle determination unit 1615 receives the lane information recognized by the lane recognition unit 1611 and information about the obstacle recognized by the obstacle recognition unit 1613, and determines whether the obstacle exists in the driving lane.

That is, when no obstacle exists in the lane recognized by the lane recognition unit 1611, the driving lane obstacle determiner 1615 determines that the driving lane obstacle is not a driving lane obstacle. In this case, the caution warning unit 1700 may not need the control for the caution warning.

However, when the obstacle recognized by the obstacle division unit 1615 is a pedestrian rather than a vehicle, the driving lane obstacle determination unit 1615 may determine whether the driving lane obstacle determination unit 1615 is not a driving lane obstacle. 1700 may perform control for the attention path. This is because pedestrians can move out of line with the lane.

Attention warning unit (1700) is a driver to prevent collision between vehicles or vehicles and pedestrians. Warnings may be given to drivers or pedestrians of other vehicles.

Examples of the caution warning performed by the caution warning unit 1700 may include a warning sound, a screen display, a seat belt vibration, or an automatic brake, but are not limited thereto. That is, the attention warning unit 1700 may perform the function of the determination sign output unit 1628 of the sign recognition unit to be described later.

The sign recognition unit 1620 includes a sign candidate selection unit 1622, a second database unit 1624, a sign determination unit 1626, and / or a determination sign output unit 1628.

The sign candidate selecting unit 1622 may select a candidate of the sign from the 3D volume image generated by the 3D preprocessor 1400. Specifically, the sign candidate selection unit 1622 may include the 3D preprocessor 1400 before the 2D / 3D matching unit 1500 extracts an image of the ROI using the 2D image and the 3D volume image to determine the sign. The candidate of the sign can be selected from the three-dimensional volume image generated by.

More specifically, in the case of a sign, the sign candidate selection unit 1622 may be a sign candidate when the sign is a metallic object and the intensity value is greater than or equal to a predetermined value.

The sign may be a traffic sign, but is not limited thereto.

The sign determination unit 1626 extracts a sign part from a sign candidate image selected from a 3D volume image or a region of interest extracted from the sign candidate image using a line edge extraction method.

4 is a diagram illustrating an example of a method of extracting a sign part from an image by the sign determining unit of the safe driving support apparatus according to the present invention.

Referring to FIG. 4, the signs extracted by the sign determination unit 1626 generally have the shape of a square, a triangle, and a circle. Accordingly, the sign determiner 1626 extracts line edges forming angles of 0 degrees, 45 degrees, 90 degrees, and 135 degrees using a convolution operation. If the convolution operation value is higher than a preset threshold, the sign determination unit 1626 may determine that the sign candidate image or the ROI image extracted from the sign candidate image has a strong line component.

In this case, the sign determination unit 1626 may determine the type of sign by matching the image formed based on the extracted line component with the sign image stored in the second database unit 1624.

The second database unit 1624 stores images of various signs.

5 is a view for explaining an example of a sign image stored in a second database unit of the safe driving support apparatus according to the present invention.

The image of the signs stored in the second database unit 1624 is not limited to FIG. 5, and may store images of all signs currently present.

The sign determined by the sign determiner 1626 may be output on a display provided in the vehicle, and the output image may be an image formed based on the extracted line component or a sign image stored in the matched second database unit 1624. Can be.

The safe driving support apparatus 100 according to the present invention may distinguish a vehicle in a driving lane from a vehicle in another lane and perform a collision avoidance function.

In addition, the safe driving support apparatus 100 according to the present invention can recognize a pedestrian as well as a vehicle, thereby preventing a collision with a pedestrian.

The safe driving support apparatus 100 according to the present invention may recognize a sign with higher accuracy and grasp the type of the sign.

6 is a flowchart illustrating a safe driving support method according to an embodiment of the present invention.

Referring to FIG. 6, the camera information receiver 1100 receives camera information that is 2D image information (step S605).

The camera information received by the camera information receiver 1100 may be changed in size and format by the image signal processor 1300, and image quality improvement may be performed at step S610.

The rider information receiving unit 1200 receives rider information collected from the rider (S615).

The 3D preprocessor 1400 receives the information collected by the rider from the rider information receiver 1200, generates intensity information and depth information, and generates a 3D volume image using the generated intensity information and depth information (S620). ).

The lane recognizing unit 1611 of the obstacle classification recognizing unit 1610 may recognize the driving lane of the vehicle using the camera information (step S625).

The sign candidate selecting unit 1622 may select an image that is likely to be an image including a sign from the generated 3D volume image (S630).

The 2D / 3D matching unit 1500 extracts a region of interest using camera information and 3D volume images, which are 2D image information (S635). The 2D / 3D matching unit 1500 may extract an ROI from the image selected by the sign candidate selecting unit 1622. The region of interest may mean an area that needs to be identified as an obstacle.

The sign determination unit 1626 extracts a sign shape from the ROI image and compares the sign shape with the image stored in the second database unit 1624 where various sign images are stored (S640).

The determination sign output unit 1628 may output the sign information determined by the sign determination unit 1626 on the display (S645).

The obstacle classification unit 1610 may recognize the existence of the obstacle using the lane information, the 3D volume image, and / or the ROI image recognized by the lane recognition unit 1611, and determine whether the obstacle exists in the driving lane. In operation S650, the obstacle may be recognized as a pedestrian or a vehicle.

The attention warning unit 1700 may warn the pedestrian, the vehicle, or the driver based on the obstacle classification recognition (S655).

It is to be understood that the block diagrams of the obstacle classification recognition apparatus 100 and the sign recognition apparatus 400 according to the preferred embodiment of the present invention represent an exemplary conceptual view embodying the principles of the present invention. Similarly, all of the flowcharts should be understood to represent various processes that may be substantially represented on a computer-readable medium and executed by a computer or processor, whether the computer or processor is explicitly shown.

The functionality of the various elements shown in the figures, including functional blocks represented by a processor or similar concept, can be provided by the use of dedicated hardware as well as hardware capable of executing software in association with appropriate software. When provided by a processor, the functions may be provided by a single dedicated processor, a single shared processor, or a plurality of individual processors, some of which may be shared.

Also, the explicit use of terms such as processor, control, or similar concepts should not be interpreted exclusively as hardware capable of running software, and may be used without limitation as a digital signal processor (DSP) (ROM), random access memory (RAM), and non-volatile memory. Other hardware may also be included.

It will be apparent to those skilled in the art that various modifications, substitutions and substitutions are possible, without departing from the scope and spirit of the invention as disclosed in the accompanying claims. will be. Accordingly, the embodiments disclosed in the present invention and the accompanying drawings are not intended to limit the technical spirit of the present invention but to describe the present invention, and the scope of the technical idea of the present invention is not limited by the embodiments and the accompanying drawings. . The scope of protection of the present invention should be construed according to the following claims, and all technical ideas within the scope of equivalents should be construed as falling within the scope of the present invention.

Claims (10)

A camera information receiver configured to receive information collected by a camera of the vehicle;
A rider information receiver configured to receive information collected by a rider; And
And an information recognizing unit for recognizing information about an object included in the received information by using the information received by the camera information receiving unit and the rider information receiving unit.
The method of claim 1,
The information recognition unit,
An obstacle classification recognition unit for recognizing information on an object which is an obstacle to driving of the vehicle among objects included in the received information;
The obstacle classification recognition unit,
A lane recognizing unit recognizing a lane including a lane in which the vehicle travels using information received by the camera information receiving unit;
An obstacle recognition unit that recognizes the obstacle by using the information received from the camera information receiver and the rider information receiver; And
And a driving lane obstacle determining unit determining whether the recognized obstacle is an obstacle present in the driving lane recognized by the lane recognition unit.
3. The method of claim 2,
The obstacle classification recognition unit,
Safe driving support device, characterized in that it further comprises an obstacle discriminating unit for determining that the recognized obstacle is any one of a pedestrian, a vehicle, or noise.
The method of claim 3, wherein
The safe driving support device,
A 3D preprocessing unit generating intensity information and depth information using the information received by the rider information receiving unit, and generating a 3D image using the generated information;
The obstacle division unit,
A first database unit for storing a plurality of pedestrian images and vehicle images,
A gradient is calculated using the camera information and the generated three-dimensional image, edge information is obtained from the image using the calculated gradient, and contour information of the recognized obstacle is obtained using the edge information. After acquiring, matching the contour information with the image stored in the first database unit to determine that the recognized obstacle is any one of pedestrian, vehicle or noise (Noise).
3. The method of claim 2,
The safe driving support device,
When the determination of the driving lane obstacle determination unit determines that the recognized obstacle is an obstacle existing between the recognized driving lanes,
And a warning device configured to perform at least one of a sound, a screen display, a seat belt vibration of the vehicle, and a brake control of the vehicle.
The method of claim 1,
The information recognition unit,
Includes a sign recognition unit for recognizing information on the sign of the objects included in the received information,
The sign recognition unit,
A sign candidate selecting unit which selects an image assumed to be a sign using information received from the camera information receiving unit and the rider information receiving unit;
A second database unit storing a plurality of sign images; And
And a sign determination unit configured to determine a sign type of the selected image by matching the image selected by the sign candidate selecting unit with the sign image stored in the second database unit.
The method according to claim 6,
The safe driving support device,
A 3D preprocessing unit generating intensity information and depth information using the information received by the rider information receiving unit, and generating a 3D image using the generated information;
The sign determination unit,
The line edge of the three-dimensional image is extracted to determine a part estimated to be a sign part, and the sign type of the selected image is determined by matching the determined sign part and the sign image stored in the second database part. Driving support device.
A camera information receiving step of receiving information collected by a camera of the vehicle;
A rider information receiving step of receiving information collected by a rider; And
And an information recognizing step of recognizing information on an object included in the received information by using the information received at the camera information receiving step and the rider information receiving step.
The method of claim 8,
The information recognition step,
An obstacle classification recognition step of recognizing information on an object which is an obstacle to driving of the vehicle among objects included in the received information;
The obstacle classification recognition step,
A lane recognition step of recognizing a lane including a lane in which the vehicle is driven by using the information received in the camera information receiving step;
An obstacle recognition step of recognizing an obstacle using information received in the camera information receiving step and the rider information receiving step; And
And a driving lane obstacle determining step of determining whether the recognized obstacle is an obstacle present in the traveling lane recognized in the lane recognition step.
The method of claim 8,
The information recognition step,
And a sign recognition step of recognizing information on a sign among objects included in the received information.
The sign recognition step,
A sign candidate selection step of selecting an image assumed to be a sign using information received in the camera information receiving step and the rider information receiving step; And
And a sign determination step of determining a sign type of the selected image by matching the image selected in the sign candidate selection step with an image stored in a second database unit in which a plurality of sign images are stored.

Images