Method and apparatus for determining a road condition
First Claim
Patent Images
1. A method of determining a road condition of a road surface of a road on which a subject vehicle is traveling, comprising:
- with a camera of the subject vehicle, acquiring successive camera images including the road surface, wherein the successive camera images include an earlier image acquired at an earlier time and a later image acquired at a later time after the earlier time;
in the later image, determining a first image area including a first surface area on the road surface;
in the earlier image, determining a second image area including a second surface area on the road surface;
performing image processing including classification of the first image area in the later image to classify the first surface area on the road surface at the later time into a first class among plural pre-determined classes respectively representing different road conditions;
performing image processing including classification of the second image area in the earlier image to classify the second surface area on the road surface at the earlier time into a second class among the plural pre-determined classes respectively representing different road conditions;
performing merging or plausibility-checking, wherein the merging comprises merging the classification of the first surface area on the road surface at the later time in the first class with the classification of the second surface area on the road surface at the earlier time in the second class to produce merged information regarding the road condition, and wherein the plausibility-checking comprises checking a plausibility of the classification of the first surface area on the road surface at the later time in the first class relative to the classification of the second surface area on the road surface at the earlier time in the second class to produce plausibility-checked information regarding the road condition; and
outputting the merged information or the plausibility-checked information as output information.
3 Assignments
0 Petitions
Accused Products
Abstract
A method and an apparatus determine a road condition using a vehicle camera. At least one image is acquired using the camera. A first image area that includes an image of the road surface is determined in the at least one image. A classifier assigns the first image area to at least one class (among pre-established classes) that represents a specific road condition of the road surface. Information regarding this specific road condition is output.
37 Citations
20 Claims
-
1. A method of determining a road condition of a road surface of a road on which a subject vehicle is traveling, comprising:
-
with a camera of the subject vehicle, acquiring successive camera images including the road surface, wherein the successive camera images include an earlier image acquired at an earlier time and a later image acquired at a later time after the earlier time; in the later image, determining a first image area including a first surface area on the road surface; in the earlier image, determining a second image area including a second surface area on the road surface; performing image processing including classification of the first image area in the later image to classify the first surface area on the road surface at the later time into a first class among plural pre-determined classes respectively representing different road conditions; performing image processing including classification of the second image area in the earlier image to classify the second surface area on the road surface at the earlier time into a second class among the plural pre-determined classes respectively representing different road conditions; performing merging or plausibility-checking, wherein the merging comprises merging the classification of the first surface area on the road surface at the later time in the first class with the classification of the second surface area on the road surface at the earlier time in the second class to produce merged information regarding the road condition, and wherein the plausibility-checking comprises checking a plausibility of the classification of the first surface area on the road surface at the later time in the first class relative to the classification of the second surface area on the road surface at the earlier time in the second class to produce plausibility-checked information regarding the road condition; and outputting the merged information or the plausibility-checked information as output information. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16)
-
-
17. A method of determining a road condition using a vehicle camera of a subject vehicle, comprising:
-
acquiring plural images with the vehicle camera; determining, in the images, a first image area that includes an image of a road surface; providing the first image area to a classifier, and with the classifier assigning the first image area to at least a first road condition class among plural road condition classes that respectively represent different road conditions; determining, in the images, a second image area that includes an image of a section of the road surface that lies farther ahead than the first image area; providing the second image area to the classifier, and with the classifier assigning the second image area to at least a second road condition class among the plural road condition classes; checking a plausibility of the assignment of the first image area of a currently acquired one of the images to at least the first road condition class, relative to the assignment of the second image area of a previously acquired one of the images to at least the second road condition class; and outputting information regarding at least one plausibility-checked road condition based on a result of the checking of the plausibility. - View Dependent Claims (18)
-
-
19. A method of determining a road condition using a vehicle camera of a subject vehicle, comprising:
-
acquiring plural images with the vehicle camera; determining, in the images, a first image area that includes an image of a road surface; providing the first image area to a classifier, and with the classifier assigning the first image area to at least a first road condition class among plural road condition classes that respectively represent different road conditions; determining, in the images, a second image area that includes an image of a section of the road surface that lies farther ahead than the first image area; providing the second image area to the classifier, and with the classifier assigning the second image area to at least a second road condition class among the plural road condition classes; merging the assignment of the first image area of a currently acquired one of the images to at least the first road condition class, with the assignment of the second image area of a previously acquired one of the images to at least the second road condition class; and outputting information regarding at least one merged road condition based on a result of the merging. - View Dependent Claims (20)
-
Specification