Method of analyzing sequences of road images, device for implementing it and its application to detecting obstacles

• US 5,706,355 A
• Filed: 09/19/1996
• Issued: 01/06/1998
• Est. Priority Date: 03/22/1991
• Status: Expired due to Term

• Current Assignee: Thomson-CSF
• Original Assignee: Thomson-CSF
• Inventors: Even, Gilles, Raboisson, Stephane
• Primary Examiner(s): Couso, Jose L.
• Assistant Examiner(s): Do, Anh Hong
• Application Number:
  US08/715,248
• Time in Patent Office: 474 Days
• Field of Search: 382/100, 382/190, 382/215, 382/104, 382/272, 364/436, 395/131
• US Class Current: 382/104
• CPC Class Codes: G01S 13/867 Combination of radar systems with cameras
• CPC Class Codes: G05D 1/0242 using non-visible light signals, e.g. IR or UV signals
• CPC Class Codes: G05D 1/0246 using a video camera in combination with image processing means
• CPC Class Codes: G05D 1/0257 using a radar radar systems designed for anti-collision purposes between land vehicles or between land vehicle and fixed obstacles G01S13/931
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/30256 Lane; Road marking
• CPC Class Codes: G06T 2207/30261 Obstacle
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G08G 1/16 Anti-collision systems road vehicle drive control systems for predicting or avoiding probable or impending collision otherwise than by control of a particular sub-unit B60W30/08

Abstract

A method and system of analyzing sequences of road images taken by a camera carried on board of a vehicle. A luminance signal of the road images from which contours of the road are extracted is generated. An analysis of color video signals based on the fact that the colorimetric components of a road region in an image do not include a green or red dominant is then combined with the luminance signal processing, in order to confirm, with good reliability, the limits of the road in the image on the basis of the contours. A mask of the road is then established. This method and system are applicable, in particular, to detecting obstacles on the road, and the information which it generates may be combined with information originating from other detectors in an obstacle detection system applicable to motor vehicle traffic.

Claims

1. A method of analyzing sequences of image video signals, taken by a camera on board a vehicle, for extracting a contour of a road from the image video signals, comprising the steps of:

analyzing luminance information of the image video signals to extract contours and regions from the image video signals to generate a mask signal of the road;

analyzing color information of the image video signals comprising the substeps of;

determining average blue, red and green components for points in the image video signals;

comparing the average blue components with the average red and green components for the points in the image video signals; and

determining points at which a deviation between the average blue components and the average red or green components in the image video signals exceeds a first predetermined threshold; and

combining the mask signal of the road from the analyzed luminance information with the determined points from the analyzed color information to extract the contour of the road.

2. The method according to claim 1, wherein the step of analyzing the luminance information further comprising the substeps of:

calculating gradients of the luminance information of the image video signals, the gradients having amplitudes;

forming an image of contours formed of a plurality of image points from the amplitudes of the gradients of the luminance information;

generating energy values of the energy of each of the plurality of image points based on the image of contours;

thresholding the energy values of each of the plurality of image points by comparing the energy values with a second predetermined threshold, and retaining only the energy values which exceed the second predetermined threshold, the mask signal of the road being generated based on the retained energy values.

3. The method according to claim 1, wherein the steps of analyzing the luminance information and analyzing the chrominance information are executed in parallel.

4. The method according to claim 1, wherein the steps of analyzing the luminance information and analyzing the chrominance information are executed in parallel.

5. A system of analyzing sequences of image video signals, taken by a camera on board a vehicle, for extracting a contour of a road from the image video signals, comprising:

first analyzing means for analyzing luminance information of the image video signals to extract contours and regions from the image video signals to generate a mask signal of the road;

second analyzing means for analyzing color information of the image video signals, the second analyzing means comprising,

means for determining average blue, red and green components for points in the image video signals,

means for comparing the average blue components with the average red and green components for the points in the image video signals, and

means for determining points at which a deviation between the average blue components and the average red or green components in the image video signals exceeds a first predetermined threshold; and

combining means for combining the mask signal of the road from the analyzed luminance information with the determined points from the analyzed color information to extract the contour of the road.

6. A system according to claim 5, wherein the analyzing means for analyzing the luminance information further comprises:

calculating means for calculating gradients of the luminance information of the image video signals, the gradients having amplitudes, the amplitudes of the gradients constituting an image of contours formed of a plurality of image points from the amplitudes of the gradients of the luminance information, and for generating energy values of the energy of each of the plurality of image points based on the image of contours; and

thresholding means for thresholding the energy values of each of the plurality of image points by comparing the energy values with a second predetermined threshold, and retaining only the energy values which exceed the second predetermined threshold, the mask signal of the road being generated based on the retained energy values.

7. The method according to claim 6, wherein the first analyzing means for analyzing the luminance information and the second analyzing means for analyzing the chrominance information operate in parallel.

8. The method according to claim 5, wherein the first analyzing means for analyzing the luminance information and the second analyzing means for analyzing the chrominance information operate in parallel.