KR20170020961A

KR20170020961A - Lane recognizion device in a vehicle and control method of thereof

Info

Publication number: KR20170020961A
Application number: KR1020150115104A
Authority: KR
Inventors: 조민관
Original assignee: 주식회사 만도
Priority date: 2015-08-17
Filing date: 2015-08-17
Publication date: 2017-02-27

Abstract

A lane recognition device and a lane recognition method are disclosed. A lane recognizing apparatus according to an embodiment of the present invention includes a photographing unit for photographing a road image on which a vehicle travels, a sun sensing unit for sensing the position of the sun, The lane by the shadow of the median of the road image is excluded from among the extracted lane candidates based on the sun position sensed by the sun sensing unit and the remaining except the lane by the shadow of the median separator is recognized as the lane And a lane recognition unit.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a lane recognition apparatus and a lane recognition method,

The present invention relates to a lane recognition apparatus and a lane recognition method, and more particularly, to a lane recognition apparatus and a lane recognition method for recognizing lanes in a shadow environment.

A lane keeping system such as a lane keeping assistance system (Lane Keeping Assistance System) and a lane departure warning system recognizes a lane based on lane information of a road image photographed using a front camera. At this time, in order for the lane-keeping system to accurately recognize the lane, the detected lane information must be accurate.

However, there is a problem that the lane recognition performance deteriorates when shadows are generated on the road from the median separator of the road. This is because the shadows are located above the lane and change the general color and brightness of the lane.

In addition, when the brightness of the boundary between the light passing through the median separator and the shadow of the median separates suddenly, and the edge extraction process used for lane recognition has an output similar to the lane, .

According to an embodiment of the present invention, there is provided a lane recognition apparatus and a lane recognition method capable of more accurately and reliably recognizing a lane under a shadow environment in which a shadow of a median separator can be misinterpreted as a lane.

According to an aspect of the present invention, there is provided an information processing apparatus comprising: a photographing unit photographing a road image on which a vehicle runs; A sun sensing unit for sensing the position of the sun; And a lane-by-shadow lane of the median of the road image from among the extracted lane candidates based on the sun position sensed by the sun sensor, using edge detection in the road image photographed through the photographing unit, And a lane recognition unit for recognizing the remaining lanes except the lane by the shadow of the median separator as a lane.

The lane recognizing unit may include estimating a shadow direction and a shadow boundary of the median using the position and brightness of the sun and the position and height of the median.

The lane recognizing unit may include estimating a lane by the shadow of the median based on the shadow direction and the shadow boundary of the median.

The lane recognition unit may further generate an image emphasizing at least one of yellow or blue in the road image photographed through the photographing unit, and extracting the lane candidate based on the generated image.

According to another aspect of the present invention, there is provided an image processing method for shooting an image of a road on which a vehicle travels through a photographing unit, converting an image of the road into a gray image form of the photographed road image and detecting an edge, A lane candidate corresponding to the shadow of the median of the road image among the extracted lane candidates based on the sensed sun position, recognizes a lane candidate corresponding to a shadow of the median of the road image, And recognizing the remainder as a lane can be provided.

It is also possible to estimate the shadow direction and shadow boundaries of the median using the position and brightness of the sun, the position and height of the median, and estimate the shadow direction and shadow boundary of the median based on the shadow direction and shadow boundaries of the median And recognizing lane candidates corresponding to shadows.

The lane candidate extraction may further include generating an image in which at least one of yellow or blue is emphasized in the road image photographed through the photographing unit, and extracting the lane candidate based on the generated image.

The embodiments of the present invention can prevent the shadow of the median separator from being mistaken as a lane by using the sun position and the sun brightness, so that the lane can be recognized more accurately and reliably.

1 is a control block diagram of a lane recognition apparatus according to an embodiment of the present invention.
2 is a control block diagram of a lane recognition unit of a lane recognition apparatus according to an embodiment of the present invention.
3 is a view for explaining a situation in which a shadow of a median separator can be misinterpreted as a lane on a road image photographed by a lane recognizing apparatus according to an embodiment of the present invention.
FIG. 4 is a diagram for explaining how a lane marker is extracted in a shadow situation in which a shadow of a median separator in a lane recognizing device according to an embodiment of the present invention is applied to a lane.
5 is a control flowchart for explaining a lane recognition process in the lane recognition apparatus according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. The embodiments described below are provided by way of example so that those skilled in the art will be able to fully understand the spirit of the present invention. The present invention is not limited to the embodiments described below, but may be embodied in other forms. In order to clearly explain the present invention, parts not related to the description are omitted from the drawings, and the width, length, thickness, etc. of the components may be exaggerated for convenience. Like reference numerals designate like elements throughout the specification.

1 is a control block diagram of a lane recognition apparatus according to an embodiment of the present invention.

Referring to FIG. 1, the lane recognizing apparatus includes a lane recognizing section 10 for performing overall control.

The lane recognizing section 10 is electrically connected to the photographing section 20 and the sun sensing section 21.

The photographing section 10 photographs roads, for example, roads ahead of the vehicle, and generates road images. The photographing unit 10 can photograph a road image of a vehicle using a CCD or a CMOS device. In the present embodiment, only the photographing of the road image is performed using the CCD or CMOS device provided in the photographing unit 10, but the present invention can also be realized by receiving the image from the external camera.

The sun sensing portion 21 senses the sun position. The sun sensing part 21 is made of a combination of a slit and a photodiode to sense the sun's position. The solar sensing unit 21 outputs a change in the position of the sun incident on the photodiode to an electrical signal or a digital signal. For example, in order to partially or totally block the sunlight according to the angle of the sunlight incident on the photodiode, the sun sensor 21 generates a sun sensor of a cylindrical structure having a circular opening at the top and a photodiode Converts the current into a voltage and transmits it to the lane recognition section 10 as an analog signal or a digital signal. The sun position can include sun bearing and solar altitude.

Further, the sun sensing portion 21 can sense the sun brightness together with the sun position. The sun sensing unit 21 transmits information related to the sun position and the sun brightness to the lane recognizing unit 10. The lane recognizing unit 10 recognizes the sun position and the sun brightness analyzed based on the sun position sensed through the sun sensing unit 21 and the information related to the sun brightness.

The lane recognition unit 10 recognizes a lane in a shadow region by linking the road image photographed through the photographing unit 10 with information related to the sun position and brightness sensed through the sun sensing unit 21. The performance of the recognition of the surroundings of the vehicle calculated based on the output of the photographing unit 10 and the output of the sun sensing unit 21 is more effective in terms of misrecognition and robustness than that of the individual system.

The lane recognition unit 10 extracts a lane candidate from the road image photographed through the photographing unit 10, and based on the sun position and brightness sensed through the sun sensing unit 21, Except for the lane by the shadow of the median separator, except for the lane by the shadow of the median separator.

The lane recognition unit 10 measures the direction of sunlight with the output of the sun sensing unit 21 to determine whether a shadow direction and a shadow are generated. To distinguish the shadow interface from the lane, use the shadow direction and shadow creation information. The lane recognition unit 10 uses the sun position and brightness information among the lane candidates extracted from the road image photographed through the photographing unit 20, and excludes the lane candidate determined as a shadow from the lane candidate. The lane recognition unit 10 predicts the shadow of the median separator and identifies whether or not the median separator exists to distinguish it from the boundary lane of the shadow.

In addition, the lane recognition unit 10 can predict the range and concentration of shadows using the sun position and brightness obtained from the output of the sun sensing unit 21. The lane recognition unit 10 generates a yellow emphasized image in the road image photographed through the photographing unit 10 according to the shadow prediction result so that the lane boundary can be easily distinguished according to the range of shadows, The color and boundary contrast can be improved by generating an image or generating an image in which both colors are emphasized and performing a lane candidate extraction based on the generated image.

The lane recognition section 10 includes, for example, a lane candidate extraction part 11, a median separation detection part 12, a shadow estimation part 13, a lane candidate selection part 14 and a lane recognition part 15 can do.

The lane candidate extracting part 11 converts a road image of RGB into a gray image, detects an edge using the pixel transform in the converted gray image, and extracts the lane candidate using the detected edge.

FIG. 3 is a view for explaining a situation in which a shadow of a median separator can be misinterpreted as a lane in a road image photographed by a lane recognizing apparatus according to an embodiment of the present invention. FIG. Fig. 3 is a diagram for explaining how to extract a lane candidate in a shadow situation in which a shadow of a median separator is displayed on a lane in a lane recognizing device.

3 and 4, the lane candidate extracting part 11 detects lane candidates L1, L1 ', L2 and L3 using the edge detection in the road image converted into the gray image form of FIG. do. As can be seen from the human's view, L1 'is the shadow of the median separator (G), which is mistaken for the lane as the brightness of the median separator (G) changes abruptly.

Referring again to FIG. 2, the median separator detecting part 12 detects the median separator in the road image converted into the gray image.

The shadow estimation part (13) estimates the shadow direction and shadow generation of the median based on the sun position. At this time, the shadow estimation part 13 can estimate the shadow direction and shadow generation of the median based on the sun position and brightness.

Generally, if the altitude of the sun is high, the length of the shadow is short, and if the altitude of the sun is low, the length of the shadow is long.

The lane candidate selection part 14 calculates the shadow length along the height of the median separator in the shadow direction at the current sun position among the extracted lane candidates, recognizes the lane candidate corresponding to the shadow of the median separator, Select the rest except the lane candidate that corresponds to the shadow of the separator.

The lane recognition part 15 recognizes the lane candidate selected by the lane candidate selecting part 14 as a lane.

5 is a control flowchart for explaining a lane recognition process in the lane recognition apparatus according to an embodiment of the present invention.

Referring to FIG. 5, first, the lane recognition unit 10 photographs a road on which a vehicle travels through a photographing unit 20 so as to acquire a road image of a road on which the vehicle travels (100).

After taking a road image, the lane recognition unit 10 converts a road image photographed through the photographing unit 20 into a gray image, and recognizes a pixel having a difference between pixels of the gray image equal to or larger than a predetermined value as an edge point. An edge is detected in the image (102). That is, a point where a pixel value changes rapidly in a road image is recognized as an edge point.

After edge detection, the lane recognition unit 10 detects a lane candidate based on the detected edge (104).

After detecting the lane candidate, the median separator is detected in the road image (106). The median separator detection is performed in a known manner using edges detected for the road image converted to a gray image. At this time, when the median separator is detected, the position and height of the median separator can be recognized.

After detection of the median separator, the lane recognition unit 10 senses the sun position and brightness through the sun sensing unit ().

After detecting the sun position and brightness, the lane recognition unit 10 determines whether a shadow direction and a shadow are generated using the sun position, brightness, and whether the center separator is detected (110).

After determining whether the shadow direction and the shadow are generated, the lane recognizing unit 10 estimates the shadow boundary of the median separator using the position of the sun and the position and height of the median separator, Among the candidates, the lane candidate corresponding to the shadow of the median is recognized, and the remaining lane candidates except the lane candidate corresponding to the shadow of the recognized median are selected as the lane candidate (112).

After selecting the lane candidate, the lane recognition unit 10 recognizes the selected lane candidate as a lane (114).

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. It will be understood that various modifications and changes may be made without departing from the scope of the appended claims.

10: lane recognition part 11: lane candidate extraction part
12: Median separation detection part 13: Shadow estimation part
14: Lane candidate selection part 15: Lane recognition part
20: photographing part 21: sun sensing part

Claims

A photographing unit photographing a road image on which the vehicle travels;
A sun sensing unit for sensing the position of the sun; And
A lane by the shadow of the median of the road image is extracted from the extracted lane candidates based on the sun position sensed by the sun sensor, And a lane recognition unit which recognizes the lane except for the lane by the shadow of the median separator as a lane.

The method according to claim 1,
Wherein the lane recognition unit includes estimating a shadow direction and a shadow boundary of the median using the position and brightness of the sun and the position and height of the median.

3. The method of claim 2,
Wherein the lane recognizing unit estimates a lane based on the shadow of the median separator based on a shadow direction and a shadow boundary of the median separator.

The method according to claim 1,
Wherein the lane recognition unit further generates an image emphasizing at least one of yellow or blue in the road image photographed through the photographing unit and extracts the lane candidate based on the generated image.

The image of the road on which the vehicle travels is photographed through the photographing unit,
An edge is detected after converting the photographed road image into a gray image form,
Extracts a lane candidate using the detected edge,
Detecting the position of the sun through the sun sensing part,
Recognizes a lane candidate corresponding to a shadow of the median of the road image from among the extracted lane candidates based on the sensed sun position,
And recognizing the remainder excluding the recognized lane candidate as a lane.

6. The method of claim 5,
Estimating a shadow direction and a shadow boundary of the median using the position and brightness of the sun, the position and height of the median, and determining a shadow direction and a shadow boundary of the median based on the shadow direction and the shadow boundary of the median And recognizing the corresponding lane candidate.

6. The method of claim 5,
Wherein the lane candidate extraction further includes generating an image in which at least one of yellow or blue is emphasized in the road image photographed through the photographing unit and extracting the lane candidate based on the generated image.