US 2020/0394422 A1Published application

Lane detection apparatus and method and electronic device

AssigneeFujitsu Limited

InventorsXiaoyan BU, Zejing HAN

Life of this patent

Estimating the term…

Loading patent…

Abstract

Embodiments of this disclosure provide a lane detection apparatus and method and an electronic device. First, preprocessing is performed based on semantic segmentation to remove interference objects in a binary image, which may improve accuracy of a lane line detection and may be applicable to various road scenarios, and the result based on semantic segmentation may automatically extract a lane line region image containing one or more lane lines, thereby automatically performing perspective transformation to perform search and fitting on the one or more lane lines, including achieving multi-lane detection. And furthermore, by synthesizing detection results of a plurality of input images, accuracy and integrity of the lane detection may further be improved.

Claims10

Dependent claims

An apparatus, comprising:

a processor to couple to a memory and to,
detect a plurality of input images to obtain respective lane line detection results indicating one or more lane lines in the input images;
and
obtain a synthetical lane line detection result according to the obtained one or more lane lines in the input images corresponding to the respective lane line detection results,
wherein to obtain the one or more lane lines of an input image among the input images, the processor is to,
determine regions of objects of various types in the input image based on a semantic segmentation;
remove interference objects from a binary image of the input image according to the regions of objects of various types in the input image, to obtain a preprocessed binary image;
extract a lane region image from the preprocessed binary image according to the regions of objects of various types in the input image;
perform a perspective transformation on the extracted lane region image to obtain an overhead-view image;
perform a search of one or more lane lines and a fitting on the overhead-view image to determine one or more lane lines in the overhead-view image;
and
perform a perspective inverse transformation on the overhead-view image with the one or more lane lines in the overhead-view image, to obtain the one or more lane lines in the input image.

9 more elements

Dependencies are inferred from the claim text.

Facts

Application: 16/887,401Filed 29 May 2020
Priority: CN 201910513841.514 Jun 2019
Published: 17 Dec 2020
Term ends: Not estimatedThe source records don’t support an estimate.

Classification5 codes

G06V 20/588
Exterior to a vehicle by using sensors mounted on the vehicle: recognition of the road, e.g. of lane markings; Recognition of the vehicle driving…
G06V 10/22
Image preprocessing: by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide…
G06V 10/50
Extraction of image or video features: by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients…
G06V 10/247
Aligning, centring, orientation detection or correction of the image: by affine transforms, e.g. correction due to perspective effects;…

Family

US 11,126,867 B2Related patent
US 2020/0394422 A1This patent

From the US filing history. Foreign family members aren’t listed.

Timeline

Citations

Open elsewhere

Building an agent or pulling many patents? Use the API or MCP server instead of scraping pages. Get an API key