Lane level traffic

Patent metadata

Field	Value
Publication number	US-9747505-B2
Application number	US-201414324797-A
Country	US
Kind code	B2
Filing date	Jul 7, 2014
Priority date	Jul 7, 2014
Publication date	Aug 29, 2017
Grant date	Aug 29, 2017

Abstract

Lane level traffic levels are determined based on traffic camera images. A controller aligns a three-dimensional map with a traffic camera view, and identifies multiple lanes in the traffic camera view based on lane delineations of the three-dimensional map. The controller calculates a traffic parameter based on the multiple lanes in image frames from the traffic camera view and provides a traffic graphic based on the traffic parameter.

First claim

We claim:

1. A method comprising:
identifying a position of a camera for a traffic camera view;
identifying a viewable distance range of the camera;
calculating, using a processor, a first distance to a first lane of a plurality of lanes;
determining a first image location in the traffic camera view for the first lane based on the first distance and the viewable distance range to the position of the camera;
calculating, using the processor, a second distance to a second lane of the plurality of lanes;
determining a second image location in the traffic camera view for the second lane based on the second distance and the viewable distance range to the position of the camera;
aligning, using the processor, a three-dimensional map with the traffic camera view;
identifying the plurality of lanes in the traffic camera view based on lane delineations of the three-dimensional map and the first distance and the second distance;
determining a frame rate for the traffic camera view;
comparing the frame rate to a threshold; and
generating a message for a traffic graphic if the frame is less than the threshold.

2. The method of claim 1 , wherein the traffic graphic parameter includes a traffic density for a number of vehicles per unit length of the plurality of lanes.

3. An apparatus comprising:
at least one processor; and
at least one memory including computer program code for one or more programs; the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to at least perform:
receive a three-dimensional point cloud including optical distance data;
align a traffic camera view with the three-dimensional point cloud, wherein the three-dimensional point cloud includes locations of multiple lanes of a road, wherein the traffic camera view is aligned with the three-dimensional point cloud based on a focal distance of a camera;
identify multiple lanes in image frames of the traffic camera view based on the locations of multiple lanes in the map data;
track vehicles in the traffic camera view according to the multiple lanes in the image frames; and
calculate a traffic graphic based on vehicles in the image frames from the traffic camera view and a frame rate for the traffic camera view.

4. The apparatus of claim 3 , wherein the traffic graphic is added to a user interface including the map data.

5. The apparatus of claim 3 , the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to at least perform:
modifying an illustration of the road based on the traffic graphic.

6. The apparatus of claim 3 , wherein the traffic camera view is aligned with map data based on a predetermined dimension of an object in the traffic camera view.

7. The apparatus of claim 3 , wherein the traffic parameter includes traffic speed, traffic flow, or traffic density.

8. The apparatus of claim 3 , wherein the vehicles are counted in the image frames using background subtraction.

Assignees

• Here Global Bv

Inventors

• Chen Xin
• Ma Xiang
• Ostrovskiy Roman
• Zhao Terence

Classifications

• G06T17/05

  Geographic models · CPC title

• G06K9/00785

  Physics · mapped topic

• G06K9/00791 Primary

  Physics · mapped topic

• G08G1/0104

  Measuring and analyzing of parameters relative to traffic conditions · CPC title

• G06V20/647

  by matching two-dimensional images to three-dimensional objects · CPC title