Method and apparatus for identifying driving lane
AssigneeSamsung Electronics Co., Ltd.
InventorsNahyup KANG, KeeChang LEE, Inwoo HA
Life of this patent
Estimating the term…
Abstract
A method and apparatus for identifying a driving lane extracts a road marking from an input image, generate a multi-virtual lane using the road marking and a segmentation image obtained by segmenting the input image into objects included in the input image based on a semantic unit, and identifies a driving lane of a vehicle based on a relative location of the driving lane in the multi-virtual lane.
Claims38
A method of identifying a driving lane, comprising:
- obtaining an input image;
- extracting a road marking from the input image;
- segmenting the input image into objects included in the input image based on a semantic unit to obtain a segmentation image;
- generating a multi-virtual lane using the road marking and the segmentation image;
- and
- identifying a driving lane of a vehicle based on a relative location of the driving lane in the multi-virtual lane.
A method of identifying a driving lane, comprising:
- extracting, from an input image, a left lane boundary line and a right lane boundary line of a driving lane of a vehicle;
- generating a segmentation image by segmenting the input image into objects included in the input image based on a semantic unit;
- generating a multi-virtual lane by fitting, in the segmentation image, the left lane boundary line and the right lane boundary line on a left side and a right side at equidistance intervals;
- determining a number of lanes of the multi-virtual lane based on whether the multi-virtual lane corresponds to a road component in the segmentation image;
- and
- identifying the driving lane by determining a relative location of the driving lane on a road on which the vehicle is traveling based on the determined number of the lanes of the multi-virtual lane.
An apparatus for identifying a driving lane, comprising:
- an image sensor configured to capture an input image;
- and
- a processor configured to extract a road marking from the input image, to obtain a segmentation image by segmenting the input image into objects included in the input image based on a semantic unit, to generate a multi-virtual lane using the road marking and the segmentation image, and to identify a driving lane of a vehicle based on a relative location in the multi-virtual lane.
A navigation apparatus comprising:
- a sensor configured to capture an image in front of a vehicle;
- and
- a processor configured to:
- generate a segmentation image by segmenting the image into portions corresponding to objects included in the input image,
- identify a left lane boundary line and a right lane boundary line from the image,
- generate virtual lines based on fitting the left lane boundary line and the right lane boundary line in the segmentation image,
- generate a multi-virtual lane based on arranging the virtual lines in parallel on at equidistant intervals in the segmentation image,
- determine a number of lanes of the multi-virtual lane based on whether the multi-virtual lane corresponds to a road object in the segmentation image, and
- identify a driving lane of the vehicle based on a relative location of the driving lane among the determined number of the lanes of the multi-virtual lane.
A vehicle comprising:
- a sensor configured to capture an image;
- and
- a driving lane identifying processor configured to extract a road marking from the image, to obtain a segmentation image by segmenting the image into objects based on a semantic unit, to generate a multi-virtual lane using the road marking and the segmentation image, and to identify a driving lane based on a relative location of the vehicle in the multi-virtual lane.
Dependencies are inferred from the claim text.