Monocular computer vision aided road vehicle driving for safety
First Claim
Patent Images
1. A versatile monocular computer vision method of processing and analyzing a sequence of monocular road/scene images captured by a single monocular image capture device for aiding road vehicle driving;
- said method comprising the steps of;
i. by processing and analyzing the monocular road/scene images, detecting a distance to a front obstacle or front vehicle;
ii. by processing and analyzing the monocular road/scene images, calculating a vehicle direction relative to a lane direction;
iii. by processing an analyzing the monocular road/scene images, determining a left/right location of said vehicle in a road lane; and
iv. estimating a speed of said vehicle, wherein said step of detecting a distance to a front obstacle or vehicle comprises the steps of;
a. from a sequence of road/scene images extracting two lane markings on both sides of said road lane and then finding an intersection point of said lane markings;
said intersection point being a vanishing point of said road lane;
b. from said vanishing point and a known focal length of a lens of a camera, finding a pitch angle and a yaw angle of said lane markings with respect to a camera coordinate system;
c. from said pitch angle and a height of said camera, calculating a distance from a location of said camera to a point which is an intersection point of an optical axis of said camera and a road plane;
d. from said images, finding horizontal lines that indicate intersections of rear wheels of front vehicles and said road plane, selecting a nearest horizontal line that is in said road lane, and judging that said horizontal line is located above or below an image center;
e. from said known camera focal length, camera height, pitch angle, and a vertical distance from said horizontal line to said image center, finding said distance from said camera location to said rear wheel of said front vehicle or obstacle.
0 Assignments
0 Petitions
Accused Products
Abstract
The present invention is a monocular computer vision technique. The technique is used to aid road vehicle driving for safety. An camera (e.g., camera or TV camera or camcorder) is installed in a vehicle facing to the road in front of the vehicle to capture a sequence of road/scene images for analyzing the driving situation. The situation includes the distance to a front obstacle or vehicle, the speed of the vehicle, and the left/right location of the vehicle in a road lane.
202 Citations
4 Claims
-
1. A versatile monocular computer vision method of processing and analyzing a sequence of monocular road/scene images captured by a single monocular image capture device for aiding road vehicle driving;
- said method comprising the steps of;
i. by processing and analyzing the monocular road/scene images, detecting a distance to a front obstacle or front vehicle;
ii. by processing and analyzing the monocular road/scene images, calculating a vehicle direction relative to a lane direction;
iii. by processing an analyzing the monocular road/scene images, determining a left/right location of said vehicle in a road lane; and
iv. estimating a speed of said vehicle, wherein said step of detecting a distance to a front obstacle or vehicle comprises the steps of;
a. from a sequence of road/scene images extracting two lane markings on both sides of said road lane and then finding an intersection point of said lane markings;
said intersection point being a vanishing point of said road lane;
b. from said vanishing point and a known focal length of a lens of a camera, finding a pitch angle and a yaw angle of said lane markings with respect to a camera coordinate system;
c. from said pitch angle and a height of said camera, calculating a distance from a location of said camera to a point which is an intersection point of an optical axis of said camera and a road plane;
d. from said images, finding horizontal lines that indicate intersections of rear wheels of front vehicles and said road plane, selecting a nearest horizontal line that is in said road lane, and judging that said horizontal line is located above or below an image center;
e. from said known camera focal length, camera height, pitch angle, and a vertical distance from said horizontal line to said image center, finding said distance from said camera location to said rear wheel of said front vehicle or obstacle. - View Dependent Claims (2)
- said method comprising the steps of;
-
3. A versatile monocular computer vision method of processing and analyzing a sequence of monocular road/scene images captured by a single monocular image capture device for aiding road vehicle driving;
- said method comprising the steps of;
i. by processing and analyzing the monocular road/scene images, detecting a distance to a front obstacle or front vehicle;
ii. by processing and analyzing the monocular road/scene images, calculating a vehicle direction relative to a lane direction;
iii. by processing an analyzing the monocular road/scene images, determining a left/right location of said vehicle in a road lane; and
iv. estimating a speed of said vehicle, wherein said left/right location of said vehicle in said road lane is computed from the ratio of two distances which are from a midpoint of a said image lower border to two intersection points of said extended image lower border and said lane markings of said road lane.
- said method comprising the steps of;
-
4. A versatile monocular computer vision method of processing and analyzing a sequence of monocular road/scene images captured by a single monocular image capture device for aiding road vehicle driving;
- said method comprising the steps of;
i. by processing and analyzing the monocular road/scene images, detecting a distance to a front obstacle or front vehicle;
ii. by processing and analyzing the monocular road/scene images, calculating a vehicle direction relative to a lane direction;
iii. by processing an analyzing the monocular road/scene images, determining a left/right location of said vehicle in a road lane; and
iv. estimating a speed of said vehicle, wherein said vehicle speed is computed based on a determination of said distance to a front vehicle;
wherein a terminal point of a dashed lane marking is first found in said sequence of road/scene images;
said distances from said camera location to said terminal point in said images are calculated respectively; and
said vehicle speed is derived from a distance difference divided by a time difference.
- said method comprising the steps of;
Specification