Road surface detection device, object detection device, object detection system, mobile object, and object detection method
An object detection device includes a processor. The processor acquires or generates a first disparity image generated on the basis of an output of a stereo camera mounted in a mobile object. In the first disparity image, pixels representing disparities are arranged on a two-dimensional plane formed by a first direction corresponding to a base-length direction of the stereo camera and a second direction intersecting the first direction. The processor approximates a relationship between a coordinate, in the second direction, of a road surface in a direction of travel of the mobile object and a disparity representing the road surface with two straight lines, the relationship being included in the first disparity image.
1. A road surface detection device comprising:
a processor configured to:
acquire or generate a first disparity map, the first disparity map being:
generated based on an output of a stereo camera that captures an image including a road surface, and
a map in which a disparity obtained from the output of the stereo camera is associated with two-dimensional coordinates formed by a first direction corresponding to a horizontal direction of the image captured by the stereo camera and a second direction intersecting the first direction, and
using two straight lines, approximate a relationship between a coordinate of the road surface in the second direction and a disparity representing the road surface, the relationship being included in the first disparity map,
wherein the two straight lines include a first straight line and a second straight line,
the processor approximates a short-distance side in a real space with respect to a side on which the stereo camera is located using the first straight line and approximates a long-distance side farther than the first straight line using the second straight line,
the processor is configured to determine coordinates of an approximation start point in accordance with the first disparity map, the approximation start point being a point at which approximation using the second straight line is started on the long-distance side relative to the first straight line, and
the processor is configured to determine the coordinates of the approximation start point based on an error between the first straight line and the disparity representing the road surface.
2. The road surface detection device according to claim 1 , wherein the processor determines the first straight line by approximating the relationship between the coordinate of the road surface in the second direction and the disparity representing the road surface from the side on which the stereo camera is located to a coordinate in the second direction corresponding to a predetermined distance using a straight line.
3. The road surface detection device according to claim 2 , wherein the processor sets the predetermined distance to a distance that is one half a distance range for detecting an object.
4. The road surface detection device according to claim 2 , wherein the processor determines a coordinate of the approximation start point in the second direction to be a coordinate corresponding to a side closer to the stereo camera in the real space than the predetermined distance.
5. The road surface detection device according to claim 1 , wherein
the processor is configured to select the second straight line from a plurality of candidate straight lines having an angle difference within a range relative to the first straight line,
the plurality of candidate straight lines are candidates of the second straight line having the approximation start point as a starting point and having angles whose angle differences from the first straight line are selected from a predetermined angle range, and
the processor is configured to select a candidate having a minimum error from the disparity representing the road surface, and to set the selected candidate as the second straight line.
6. An object detection device comprising the road surface detection device according to claim 1 , wherein the processor is configured to detect an object using the disparity representing the road surface approximated using the straight lines.
7. The object detection device according to claim 6 , wherein the processor detects the object based on a second disparity map obtained by removing the disparity representing the road surface from the first disparity map.
8. An object detection system comprising:
a stereo camera that captures a plurality of images having a disparity therebetween; and
an object detection device including at least one processor, wherein the processor is configured to:
acquire or generate a first disparity map, the first disparity map being:
generated based on an output of the stereo camera that captures an image including a road surface, and
a map in which a disparity obtained from the output of the stereo camera is associated with two-dimensional coordinates formed by a first direction corresponding to a horizontal direction of the image captured by the stereo camera and a second direction intersecting the first direction,
using two straight lines, approximate a relationship between a coordinate of the road surface in the second direction and a disparity representing the road surface, the relationship being included in the first disparity map, and
detect an object using the disparity representing the road surface approximated using the straight lines,
wherein the two straight lines include a first straight line and a second straight line,
the processor approximates a short-distance side in a real space with respect to a side on which the stereo camera is located using the first straight line and approximates a long-distance side farther than the first straight line using the second straight line,
the processor is configured to determine coordinates of an approximation start point in accordance with the first disparity map, the approximation start point being a point at which approximation using the second straight line is started on the long-distance side relative to the first straight line, and
the processor is configured to determine the coordinates of the approximation start point based on an error between the first straight line and the disparity representing the road surface.
9. A mobile object comprising the object detection system according to claim 8 .
10. An object detection method comprising:
acquiring or generating a first disparity map, the first disparity map being:
generated based on an output of a stereo camera that captures an image including a road surface, and
a map in which a disparity obtained from the output of the stereo camera is associated with two-dimensional coordinates formed by a first direction corresponding to a horizontal direction of the image captured by the stereo camera and a second direction intersecting the first direction;
using two straight lines, approximating a relationship between a coordinate of the road surface in the second direction and a disparity representing the road surface, the relationship being included in the first disparity map; and
detecting an object using the disparity representing the road surface approximated using the straight lines,
wherein the two straight lines include a first straight line and a second straight line,
the method further comprising:
approximating a short-distance side in a real space with respect to a side on which the stereo camera is located using the first straight line and approximating a long-distance side farther than the first straight line using the second straight line, and
determining coordinates of an approximation start point:
in accordance with the first disparity map, the approximation start point being a point at which approximation using the second straight line is started on the long-distance side relative to the first straight line, and
based on an error between the first straight line and the disparity representing the road surface.