KR102501234B1 - System and method for recognizing a surroundings lane using a location of structures - Google Patents

Abstract

The present invention obtains location information of structures installed on the road while a vehicle is driving on the road, and uses this to determine the presence or absence of a surrounding lane and to set a region of interest (ROI) in the surrounding lane, surrounding lane recognition using structure location information. A method and system are disclosed.
A system for recognizing surrounding lanes using structure location information according to the disclosed subject matter includes an image acquisition unit that acquires a front image by photographing the front of a vehicle; a lane detection unit for detecting an own lane or a lane in the acquired front image; a structure detector detecting a structure in the acquired frontal image; and recognizing a lane width based on the detected own lane line to set a lane ROI (Region Of Interest) in the forward image, recognizing a position of the detected structure and setting a structure ROI, and determining the lane ROI and and a control unit configured to set a region of interest (ROI) for the surrounding lane according to the ROI of the structure and recognize the surrounding lane based on the ROI.
According to the present invention, when a vehicle drives on a road, it is possible to recognize surrounding lanes using location information of a structure.

Description

System and method for recognizing a surroundings lane using a location of structures

The present invention relates to a method and system for recognizing surrounding lanes using structure location information, and more particularly, to obtain location information of structures installed on a road while a vehicle is driving on a road, and utilize the obtained location information to determine the presence or absence of surrounding lanes and interest in the surrounding lanes. The present invention relates to a method and system for recognizing a surrounding lane using structure location information, enabling a region (ROI) to be set.

In general, when a vehicle drives on a road, it recognizes a lane or an object in front by using various sensors such as an image sensor, a radar sensor, or a lidar sensor. At this time, driving lane recognition is one of the core technologies in intelligent vehicles, and has a great influence on safe driving.

However, the recognition of the surrounding lane may be affected by surrounding structures unlike the recognition of the own lane. That is, since the own lane is the driving lane, the lane display may always be inside the road structure, but the surrounding lane may or may not exist because it is not a driving lane.

In a conventional method of recognizing a neighboring lane, a region of interest (ROI) is generated using own lane information and the surrounding lane is recognized within the region of interest (ROI).

In the case of the surrounding lanes, unlike the own lanes, there are many cases where they are not on the road. Therefore, if there is a component similar to the lane where the region of interest (ROI) is formed, there are many problems in that it is misrecognized as a neighboring lane.

Republic of Korea Patent Publication No. 2012-0066220 (June 22, 2012)

An object of the present invention to solve the above problems is to acquire location information of structures installed on the road while a vehicle is driving on the road, and use this to set the presence or absence of a surrounding lane and a region of interest (ROI) in the surrounding lane It is to provide a method and system for recognizing surrounding lanes using structure location information.

A system for recognizing surrounding lanes using structure location information according to the present invention for achieving the above object includes an image acquisition unit that acquires a front image by photographing the front of a vehicle; a lane detection unit for detecting an own lane or a lane in the acquired front image; a structure detector detecting a structure in the acquired frontal image; and recognizing a lane width based on the detected own lane line to set a lane ROI (Region Of Interest) in the forward image, recognizing a position of the detected structure and setting a structure ROI, and determining the lane ROI and and a control unit configured to set a region of interest (ROI) for the surrounding lane according to the ROI of the structure and recognize the surrounding lane based on the ROI.

Also, the structure detection unit may detect a structure outside the lane or a structure outside the road based on the detected lane.

In addition, the controller may recognize the other lane as a neighboring lane when another lane exists in the surrounding lane ROI, excluding the own lane and the structure, according to the lane ROI and the structure ROI.

In addition, the structure detection unit may detect a structure having a movement change of a central point as a signpost using the speed and yaw rate of the vehicle in the obtained front image.

In addition, the control unit recognizes the position of the sign by calculating the movement of the center point according to the speed and yaw rate of the vehicle, and the principle that a lane exists around the sign using the location information of the sign Accordingly, in the front image, a predetermined area centered on the signpost is set as a peripheral lane ROI, and when other lanes other than the own lane exist in the peripheral lane ROI, the other lanes may be recognized as peripheral lanes.

In addition, the structure detection unit may distinguish between a short exposure image and a long exposure image in the front image, and detect structures that are not continuous light sources but lined up at regular intervals in the long exposure image with a line guiding facility (Delineator). there is.

In addition, the control unit sets a peripheral lane ROI centered on the gaze guiding facility in the front image using the location information of the detected gaze guiding facility, and other lanes excluding the own lane exist in the peripheral lane ROI In some cases, other lanes may be recognized as neighboring lanes.

In addition, when the detected structure exists outside the detected own lane, the control unit may recognize it as a guardrail if the number of consecutive vertical edges of the pillar portion is greater than or equal to a certain number.

The control unit may set a surrounding lane ROI using the location information of the guardrail, and recognize the other lane as a neighboring lane when another lane excluding the own lane exists in the surrounding lane ROI.

Meanwhile, a method for recognizing surrounding lanes using structure location information according to the present invention for achieving the above object includes: (a) obtaining a front image by photographing the front of a vehicle by an image acquisition unit; (b) detecting, by a lane detection unit, an own lane line or a lane line in the obtained front image; (c) detecting a structure from the acquired frontal image by a structure detection unit; (d) setting a lane ROI (Region Of Interest) in the forward image by recognizing the width of the lane based on the detected lane, and recognizing the location of the detected structure to set a structure ROI; and (e) setting a surrounding lane ROI according to the lane ROI and the structure ROI, and then recognizing the surrounding lane based on the surrounding lane ROI.

Also, in step (c), the structure detection unit may detect a structure outside the lane or a structure outside the road based on the detected lane.

In addition, in the step (e), the control unit selects another lane as a neighboring lane when another lane exists in the surrounding lane ROI, excluding the own lane and the structure, according to the lane ROI and the structure ROI. Recognizable.

Also, in the step (c), the structure detection unit may detect a structure having a movement change of a center point as a signpost using the speed and yaw rate of the vehicle in the obtained front image.

In the step (e), the control unit calculates the movement of the center point according to the speed and yaw rate of the vehicle to recognize the location of the sign, and uses the location information of the sign to determine the location of the sign. According to the principle that lanes exist in the periphery, the surrounding lane ROI is set for a certain area centered on the sign in the front image, and when other lanes other than the own lane exist in the surrounding lane ROI, the other lane is set as the surrounding lane. It can be recognized as a lane.

In addition, in the step (c), the structure detection unit distinguishes a short-exposure image and a long-exposure image in the front image, and sets a line of sight guidance facility for structures that are not continuous light sources but lined up at regular intervals in the long-exposure image ( Delineator) can be detected.

In addition, in the step (e), the control unit sets a peripheral lane ROI centered on the gaze guiding facility in the front image using the location information of the detected gaze guiding facility, and determines the own lane from the peripheral lane ROI. If there are other lanes excluded, the other lanes may be recognized as the surrounding lanes.

In addition, in the step (e), when the detected structure exists outside the detected own lane, if the number of consecutive vertical edges of the column portion exceeds a certain number, the guardrail can be recognized as

And, in the step (e), the control unit sets the ROI of the surrounding lane using the location information of the guardrail, and if there is another lane excluding the own lane in the ROI of the surrounding lane, the other lane is selected as the surrounding lane can be recognized as

According to the present invention, when a vehicle drives on a road, it is possible to recognize surrounding lanes using location information of a structure.

In addition, the surrounding lane ROI can be generated according to the location of the structure to increase the accuracy of the surrounding lane recognition, and misrecognition can be reduced because the ROI is not created.

1 is a block diagram schematically illustrating the configuration of a system for recognizing a surrounding lane using structure location information according to an embodiment of the present invention.
2 is a diagram illustrating an example of detecting lanes and structures in a forward image according to an embodiment of the present invention.
3 is a diagram illustrating an example of detecting a sign along with a lane in a forward image according to an embodiment of the present invention.
4 is a diagram illustrating an example of detecting a gaze inducing facility in a front image according to an embodiment of the present invention.
5 is a diagram illustrating an example of recognizing a guardrail based on a lane in a forward image according to an embodiment of the present invention.
6 is a flowchart illustrating a method for recognizing a surrounding lane using structure location information 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 so that those skilled in the art can easily carry out the present invention. The present invention may be embodied in many different forms and is not limited to the embodiments described herein.

In order to clearly describe the present invention, parts irrelevant to the description are omitted, and the same reference numerals are assigned to the same or similar components throughout the specification.

Throughout the specification, when a part is said to be "connected" to another part, this includes not only the case where it is "directly connected" but also the case where it is "electrically connected" with another element interposed therebetween. . In addition, when a certain component is said to "include", this means that it may further include other components without excluding other components unless otherwise stated.

When a part is referred to as being “on” another part, it may be directly on top of the other part or may have other parts in between. In contrast, when a part is said to be “directly on” another part, there are no other parts in between.

Terms such as first, second and third are used to describe, but are not limited to, various parts, components, regions, layers and/or sections. These terms are only used to distinguish one part, component, region, layer or section from another part, component, region, layer or section. Accordingly, a first part, component, region, layer or section described below may be referred to as a second part, component, region, layer or section without departing from the scope of the present invention.

The terminology used herein is only for referring to specific embodiments and is not intended to limit the present invention. As used herein, the singular forms also include the plural forms unless the phrases clearly indicate the opposite. The meaning of "comprising" as used herein specifies particular characteristics, regions, integers, steps, operations, elements and/or components, and the presence or absence of other characteristics, regions, integers, steps, operations, elements and/or components. Additions are not excluded.

Terms indicating relative space, such as “below” and “above,” may be used to more easily describe the relationship of one part to another shown in the drawings. These terms are intended to include other meanings or operations of the device in use with the meaning intended in the drawings. For example, if the device in the figures is turned over, certain parts described as being “below” other parts will be described as being “above” the other parts. Thus, the exemplary term "below" includes both directions above and below. The device may rotate 90 degrees or other angles, and terms denoting relative space are interpreted accordingly.

Although not defined differently, all terms including technical terms and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the art to which the present invention belongs. Terms defined in commonly used dictionaries are additionally interpreted as having meanings consistent with related technical literature and currently disclosed content, and are not interpreted in ideal or very formal meanings unless defined.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present invention. However, the present invention may be implemented in many different forms and is not limited to the embodiments described herein.

1 is a block diagram schematically illustrating the configuration of a system for recognizing a surrounding lane using structure location information according to an embodiment of the present invention.

Referring to FIG. 1 , the system for recognizing surrounding lanes using structure location information according to the present invention includes an image acquisition unit 110, a lane detection unit 120, a structure detection unit 130, and a control unit 140.

The image acquisition unit 110 acquires a front image by photographing the front of the vehicle with a camera.

The lane detection unit 120 detects the lane 105 including the own lane line as shown in FIG. 2 from the obtained front image. 2 is a diagram illustrating an example of detecting lanes and structures in a forward image according to an embodiment of the present invention.

The lane detection unit 120 may include a confirmation area setting module, a road marking confirmation module, a region of interest setting module, and an own lane detection module for lane detection. That is, the lane detection unit 120 may set a confirmation area including the road area of the current own lane in the front image through the confirmation area setting module, and confirm the presence of road markings in the set confirmation area through the road marking confirmation module. there is. Then, if there is a road mark, the lane detection unit 120 may calculate a difference value between the lane prediction result and the previous lane information through the region of interest setting module, set the region of interest based on the calculated difference value, and set the region of interest. Based on the region of interest set in the setting module, the lane marking may be extracted using lane tracking through the own lane detection module and recognized as the own lane. Here, lane tracking may use a prediction filter such as a Kalman filter.

Meanwhile, the lane detection unit 110 recognizes a lane by extracting lane markings. However, in the process of extracting lane markings, road markings (eg, right turn markings) may also be extracted, and thus errors may occur in extracting lane markings necessary for lane recognition. Since road markings are marked on the road like lane markings and contain the same color information as lane markings, road markings (eg, right turn markings) are also extracted in the process of extracting lane markings.

In order to eliminate the road markings that cause the above errors, the lane detection unit 110 can set a confirmation area that can include all road areas of the current own lane through the confirmation area setting module, which is the current own lane. Set to find road markings in the road area of . In order to set a confirmation area that includes all of the road areas of the current own lane, the road area of the own lane can always be included in the front image when the area of both lanes is included when the lane spans the midpoint of the image.

The structure detection unit 130 detects the structure 101 as shown in FIG. 2 in the obtained front image. Also, the structure detector 130 may detect a structure 101 outside the lane or a structure 101 outside the road based on the detected lane 105 .

In addition, the structure detector 130 may detect a structure having a movement change of a central point as a sign 102 using the speed and yaw rate of the vehicle in the obtained front image, as shown in FIG. 3 . 3 is a diagram illustrating an example of detecting a sign along with a lane in a forward image according to an embodiment of the present invention. As shown in FIG. 3 , the sign 102 continuously changes in size in the forward image. This is because a moving vehicle is approaching the sign 102 . That is, it can be confirmed that the size of the sign 102 is different when the vehicle is far from the sign 102 and when the vehicle is close to the sign 102 . Since the size of the sign 102 varies according to the driving of the vehicle, the movement of the point of the sign 102 must be calculated in order to obtain an accurate location. That is, the position of the sign 102 can be calculated by calculating the movement of the central point.

The controller 140 recognizes the width of the lane based on the detected lane and sets a lane ROI (Region Of Interest) in the front image, recognizes the location of the detected structure and sets the structure ROI, and sets the lane ROI and structure After setting the ROI of the surrounding lanes according to the ROI, the surrounding lanes are recognized based on the ROI of the surrounding lanes.

In addition, the controller 140 may recognize the other lane as a neighboring lane when another lane exists in the ROI of the surrounding lane except for the own lane and the structure according to the lane ROI and the structure ROI.

In addition, the control unit 140 recognizes the position of the sign 102 by calculating the movement of the center point according to the speed and yaw rate of the vehicle, and uses the location information of the sign 102 to determine lanes around the sign. According to this existing principle, a certain area centered on a signpost in the front image is set as the surrounding lane ROI, and when there is another lane excluding the own lane in the surrounding lane ROI, the other lane can be recognized as the surrounding lane. At this time, the controller 140 may use a motion stereo method as a method of tracking the position of the sign 102 .

In addition, the structure detection unit 130 distinguishes a short-exposure image and a long-exposure image in the front image, and induces gaze for structures that are not continuous light sources in the long-exposure image and are lined up at regular intervals as shown in FIG. 4 . It can be detected by the Delineator (103). 4 is a diagram illustrating an example of detecting a gaze inducing facility in a front image according to an embodiment of the present invention. Therefore, the controller 140 sets the ROI of the surrounding lanes around the gaze guiding facility 103 in the front image by using the location information of the detected gaze guiding facility 103, and excludes the gaze guiding facility from the set ROI of the surrounding lanes. In addition, when there is another lane other than the own lane, the other lane can be recognized as a neighboring lane.

In addition, the control unit 140, when the detected structure exists outside the detected lane 105 as shown in FIG. 5, if the number of continuous vertical edges of the pillar portion exceeds a certain number, the guardrail (Guardrail) 104 can be recognized. 5 is a diagram illustrating an example of recognizing a guardrail based on a lane in a forward image according to an embodiment of the present invention.

Accordingly, the controller 140 may set the ROI of the surrounding lane using the location information of the guardrail 104, and recognize the other lane as the surrounding lane when there is another lane excluding the own lane in the ROI of the surrounding lane. .

6 is a flowchart illustrating a method for recognizing a surrounding lane using structure location information according to an embodiment of the present invention.

Referring to FIG. 6 , in the surrounding lane recognition system 100 using structure location information according to the present invention, the image acquiring unit 110 photographs the front of the vehicle with a camera to obtain a front image (S610). At this time, the front image acquired through the image acquisition unit 110 includes the structure 101 including the lane 105 as shown in FIG. 2, or the sign (in addition to the lane 105 as shown in FIG. 3) 102) may also be included. In addition, the front image acquired through the image acquisition unit 110 may include a gaze-guiding facility 103 as shown in FIG. 4 or a guardrail 104 as shown in FIG. 5 .

Subsequently, the lane detection unit 120 detects lanes in the acquired front image (S620). That is, the lane detection unit 120 detects the lane 105 as shown in FIGS. 2 to 5 through the confirmation area setting module, the road marking confirmation module, the area of interest setting module, and the own lane detection module as described above. can At this time, the region of interest setting module may set the region of interest using lane tracking. A lane prediction result is calculated using a prediction filter that predicts lane information, and the calculated lane prediction result and the information of the previous lane are calculated. By calculating the difference value and setting the region of interest, the road marking existing in the region of interest may be removed from the region of interest. Here, the previous lane information may refer to lane information obtained when extracting a lane display prior to the currently extracted lane display. Here, the prediction filter may include a Kalman filter, and the lane prediction result and previous lane information may include information about an angle and an offset value of the lane. In addition, the own lane detection unit performs functions such as Edge Distribution Function (EDF), Steerable Filtering, Hough Transform, Inner Pixel Detection, and Inner Pixel Fitting can do.

Subsequently, the structure detection unit 130 detects a structure from the acquired frontal image (S630). That is, the structure detector 130 may detect a structure 101 outside the lane 105 or a structure 101 outside the road as shown in FIG. 2 based on the detected lane 105. there is.

In addition, the structure detector 130 may detect a structure having a movement change of a central point as a sign 102 using the speed and yaw rate of the vehicle in the obtained front image, as shown in FIG. 3 . At this time, the structure detector 130 may detect the sign 102 using the Haar-Feature-based Viola-Jones Adaboost (Adaptive Boosting) algorithm used for object detection.

In addition, in the Haar-Feature-based Viola-Jones Adaboost (Adaptive Boosting) algorithm, a plurality of strong classifiers have a cascade structure and add 1 to 2 pixel margin pixels to the image extracted from the input image. After that, it is distinguished whether it is a positive sample that has been normalized to a specific size (eg, 20 × 20), or a non-sign image (Negative Sample) randomly extracted from an image without an image.

In addition, the Haar-Feature-based Viola-Jones Adaboost (Adaptive Boosting) algorithm uses sign images (Positive Sample) and non-sign image images (Negative Sample) of the actual recognition target to train the detector (Sample Training). ) process, and scan window inspection to determine whether the sub-window input as an input is actually a 'sign' or 'non-sign' by using the scan window inspection (Sacn-Window Search) method for the region of interest (ROI) of the front image include the process

The Haar-Feature-based Viola-Jones Adaboost (Adaptive Boosting) algorithm trains the detector by setting the number ratio of the sign image (positive sample) and the non-sign image (negative sample) to 1:2.

The Haar-Feature-based Viola-Jones Adaboost (Adaptive Boosting) algorithm learns only the part consisting of 20 horizontally and 10 vertically from an image normalized to 2020 through a horizontal detector, or ), it is possible to learn only the part consisting of 10 horizontally and 20 vertically in the image normalized to 2020.

The Haar-Feature-based Viola-Jones Adaboost (Adaptive Boosting) algorithm can detect an area in which overlap occurs by a horizontal detector and a vertical detector as an actual 'sign' area.

In addition, the structure detection unit 130 distinguishes between a short exposure image and a long exposure image in the front image, and the gaze-guiding facility shown in FIG. 4 for structures lined up at regular intervals rather than continuous light sources in the long exposure image ( Delineator) (103).

Subsequently, the control unit 140 recognizes the width of the lane based on the detected lane, sets a lane ROI (Region Of Interest) in the front image, recognizes the location of the detected structure, and sets a structure ROI (S640). .

Subsequently, the controller 140 sets the ROI of the surrounding lanes according to the lane ROI and the structure ROI, and then recognizes the surrounding lanes based on the ROI of the surrounding lanes (S650).

That is, after setting the surrounding lane ROI according to the lane ROI and the structure ROI, the controller 140 excludes lanes and structures from the surrounding lane region of interest (ROI) and places other When a lane exists, another lane may be recognized as a neighboring lane.

In addition, the control unit 140 recognizes the position of the sign 102 shown in FIG. 3 by calculating the movement of the center point according to the speed and yaw rate of the vehicle, and uses the position information of the sign to determine the position of the sign. According to the principle that lanes exist in the surroundings, the ROI of the surrounding lanes is set for a certain area centered on the sign in the front image, and when there are other lanes excluding the own lane in the ROI of the surrounding lanes, the other lanes can be recognized as the surrounding lanes. can

In addition, the controller 140 sets the ROI of the surrounding lanes around the gaze guiding facility 103 in the front image by using the location information of the detected gaze guiding facility 103, and the other lanes excluding the own lane from the ROI of the surrounding lanes. When there exists, other lanes can be recognized as neighboring lanes.

The method of recognizing the gaze inducing facility 103 from the light source of the front image according to the present invention uses a different method from a general light source recognition. In the existing light source recognition, since the gaze inducing facility 103 is not a light source, it is a part that causes misrecognition. However, in the road structure, the gaze guiding facility 103 is important information. As shown in FIG. 4 , the eye guiding facility 103 is not a light source, but is helpful in recognizing the surrounding lane when the location is known in order to recognize the surrounding lane. In the existing light source recognition method, two images were used to distinguish between light sources and non-light sources. That is, a long-exposure image and a short-exposure image are used. In the case of a light source, since it appears in both long-exposure images and short-exposure images, the light sources can be recognized. However, in the case of the eye guiding facility 103, it does not appear in a single exposure. Also, in the case of gaze inducing facilities 103, they are lined up on the road. As shown in FIG. 4 , a light source and an object other than a light source can be distinguished in a long exposure and a short exposure.

In addition, the control unit 140, when the detected structure exists outside the detected lane 105 as shown in FIG. 5, if the number of continuous vertical edges of the pillar portion exceeds a certain number, the guardrail (Guardrail) 104 can be recognized.

Accordingly, the controller 140 may set the ROI of the surrounding lane using the location information of the guardrail 104, and recognize the other lane as the surrounding lane when there is a lane other than the own lane in the ROI of the surrounding lane. there is.

Guardrails are located on the center line and outside lanes. In the case of a guardrail, it is often misrecognized because its location in the image is similar to that of the surrounding lanes. Therefore, it is possible to reduce misrecognition in the surrounding lanes by distinguishing and recognizing the guardrail using the characteristics of the guardrail. As shown in Figure 5, the feature of the guardrail is at the bottom. Guardrails can be recognized by recognizing them using the continuous vertical edges of the pillars. Using this recognition information, the ROI of the surrounding lane may be set and removed.

As described above, according to the present invention, location information of structures installed on the road is obtained while the vehicle is driving on the road, and by using this, the presence or absence of the surrounding lane and the location of the structure enabling the setting of a region of interest (ROI) in the surrounding lane A method and system for recognizing surrounding lanes using information may be realized.

Those skilled in the art to which the present invention pertains should understand that the embodiments described above are illustrative in all respects and not limiting, since the present invention can be embodied in other specific forms without changing the technical spirit or essential characteristics thereof. only do The scope of the present invention is indicated by the following claims rather than the detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be construed as being included in the scope of the present invention. .

100: surrounding lane recognition system 101: structure
102: sign 103: gaze inducing facility
104: guardrail 105: lane
110: image acquisition unit 120: lane detection unit
130: structure detection unit 140: control unit

Claims (18)

An image acquisition unit for acquiring a front image by photographing the front of the vehicle;
a lane detection unit for detecting an own lane or a lane in the acquired front image;
a structure detector detecting a structure in the acquired frontal image; and
Based on the detected lane, the width of the lane is recognized and a lane ROI (Region Of Interest) is set in the forward image, a structure ROI is set by recognizing the position of the detected structure, and the lane ROI and the structure a control unit configured to set a region of interest (ROI) for a surrounding lane according to the ROI and then recognize a surrounding lane based on the ROI;
Including,
The structure detection unit,
If the structure has a movement change of the center point using the speed and yaw rate of the vehicle in the acquired front image,
A system for recognizing surrounding lanes using structure location information for detecting the structure as a sign.
According to claim 1,
The system for recognizing surrounding lanes using structure location information, wherein the structure detection unit detects a structure outside the lane or a structure outside the road based on the detected lane.
According to claim 1,
The control unit recognizes other lanes as neighboring lanes when other lanes exist in the surrounding lane region of interest (ROI) excluding the own lane and the structure according to the lane ROI and the structure ROI, using structure location information. Perimeter Lane Recognition System.
delete According to claim 1,
The control unit recognizes the position of the sign by calculating the movement of the center point according to the speed and yaw rate of the vehicle, and according to the principle that a lane exists around the sign using the location information of the sign In the front image, structure location information that is set as a peripheral lane ROI for a certain area centered on the signpost and recognizes the other lane as a peripheral lane when other lanes other than the own lane exist in the peripheral lane ROI surrounding lane recognition system.
According to claim 1,
The structure detection unit,
In the front image, a short exposure image and a long exposure image are distinguished, and in the case of a structure that is not a continuous light source in the long exposure image but is lined up at regular intervals,
A system for recognizing surrounding lanes using structure location information, which detects the structure as a delineator.
According to claim 6,
The control unit sets a peripheral lane ROI centered on the gaze guiding facility in the front image by using the location information of the detected gaze guiding facility, and when other lanes other than the own lane exist in the peripheral lane ROI, A system for recognizing other lanes as surrounding lanes using structure location information.
According to claim 1,
The control unit,
If the detected structure has more than a certain number of consecutive vertical edges of the column part,
A system for recognizing a surrounding lane using structure location information, recognizing the detected structure as a guardrail.
According to claim 8,
The control unit sets the ROI of the surrounding lane using the location information of the guardrail, and recognizes the other lane as the surrounding lane when there is a lane other than the own lane in the ROI of the surrounding lane, using structure location information Perimeter Lane Recognition System.
(a) acquiring a front image by photographing the front of the vehicle by an image acquisition unit;
(b) detecting, by an own lane detection unit, an own lane line from the obtained front image;
(c) detecting a structure from the acquired frontal image by a structure detection unit;
(d) setting a lane ROI (Region Of Interest) in the front image by recognizing the width of the lane based on the detected own lane, and recognizing the position of the detected structure to set the structure ROI ;
(e) setting a surrounding lane ROI according to the lane ROI and the structure ROI, and then recognizing the surrounding lane based on the surrounding lane ROI;
Including,
In step (c), the structure detection unit,
If the structure has a movement change of the center point using the speed and yaw rate of the vehicle in the acquired front image,
A method for recognizing surrounding lanes using structure location information for detecting the structure as a sign.
According to claim 10,
In the step (c), the structure detection unit detects a structure outside the lane or a structure outside the road based on the detected lane.
According to claim 10,
In the step (e), the control unit recognizes another lane as a neighboring lane when another lane exists in the surrounding lane region of interest (ROI) excluding the own lane and the structure according to the lane ROI and the structure ROI , A method for recognizing surrounding lanes using structure location information.
delete According to claim 10,
In the step (e), the control unit recognizes the position of the sign by calculating the movement of the center point according to the speed and yaw rate of the vehicle, and uses the position information of the sign to determine the location around the sign. According to the principle of existence of lanes, the surrounding lane ROI is set for a certain area centered on the sign in the front image, and when there are other lanes excluding the own lane in the surrounding lane ROI, the other lanes are designated as the surrounding lanes. A method for recognizing surrounding lanes using structure location information.
According to claim 10,
In the step (c), the structure detection unit,
In the front image, a short exposure image and a long exposure image are distinguished, and in the case of a structure that is not a continuous light source in the long exposure image but is lined up at regular intervals,
A method for recognizing surrounding lanes using structure location information, wherein the structure is detected by a gaze inducing facility (Delineator).
According to claim 15,
In the step (e), the control unit sets a peripheral lane ROI centered on the gaze guiding facility in the front image by using the location information of the detected gaze guiding facility, and other lanes other than the own lane are set in the peripheral lane ROI. A method for recognizing surrounding lanes using structure location information, recognizing another lane as a surrounding lane when a lane exists.
According to claim 10,
In the step (e), the control unit,
If the detected structure has more than a certain number of consecutive vertical edges of the column part,
A method for recognizing surrounding lanes using structure location information, wherein the detected structure is recognized as a guardrail.
18. The method of claim 17,
In the step (e), the control unit sets the ROI of the surrounding lane using the location information of the guardrail, and recognizes the other lane as the surrounding lane when there is a lane other than the own lane in the ROI of the surrounding lane. A method for recognizing surrounding lanes using structure location information.

Images