Method and apparatus for detection of false alarm obstacle
View Patent ↗The embodiment of the present application provides a method and apparatus for detection of false alarm obstacle, and relates to the field of identification and detection technology, in order to improve accuracy for the obstacle detection. The method includes: obtaining a first view at a moment t and a second view at a moment t−1, collected by a camera; determining location information of the same obstacle to be detected in the first view and the second view; determining motion information of the camera from moment t−1 to the moment t; judging whether the location information of the same obstacle to be detected in two views being matched with the motion information of the camera; and if not, judging the obstacle to be detected as a false alarm obstacle. The embodiment of the present application is applied to obstacle detection.
1. A method for detection of false alarm obstacle, comprising:
obtaining a first view at a moment t and a second view at a moment t−1, collected by a binocular camera, wherein the first view is a disparity map from the left view and the right view collected by the binocular camera at the moment t; and the second view is a disparity map from the left view and the right view collected by the binocular camera at the moment t−1;
determining location information of the same obstacle to be detected in the first view and the second view;
determining motion information of the binocular camera from moment t−1 to the moment t;
judging whether the location information of the same obstacle to be detected in two views being matched with the motion information of the binocular camera; and
if not, judging the obstacle to be detected as a false alarm obstacle;
wherein said determining location information of the same obstacle to be detected in the first view and the second view comprises:
specifying the first view and the second view as a target view in turn;
based on a sub-disparity value range divided from a disparity value range of the target view, setting a disparity value beyond the sub-disparity value range in the target view as a first disparity threshold, so as to generate a sub-view of the target view;
performing contour detection on the sub-view to obtain contour information of the obstacle to be detected in the sub-view; and
determining the location information of the obstacle to be detected according to the contour information of the obstacle to be detected and the disparity value of each pixel point in the sub-disparity map.
2. The method according to claim 1 , wherein,
said judging whether the location information of the same obstacle to be detected in two views being matched with the motion information of the binocular camera comprises:
determining the motion information of the obstacle to be detected according to the location information of the obstacle to be detected in the two views;
judging whether the motion information of the obstacle to be detected is consistent with the motion information of the binocular camera; and
when the motion information of the obstacle to be detected is inconsistent with the motion information of the binocular camera, judging that the location information of the same obstacle to be detected in the two views being not matched with the motion information of the camera.
3. The method according to claim 1 , wherein said judging whether the location information of the same obstacle to be detected in two views being matched with the motion information of the binocular camera comprises:
estimating estimated location information of the obstacle to be detected in the first view, according to the motion information of the binocular camera and the location information of the obstacle to be detected in the second view;
judging whether the estimated location information of the obstacle to be detected in the first view is consistent with the location information of the obstacle to be detected in the first view; and
if the estimated location information of the obstacle to be detected in the first view is inconsistent with the location information of the obstacle to be detected in the first view, judging that the location information of the obstacle to be detected in the two views being not matched with the motion information of the binocular camera.
4. The method according to claim 2 , wherein
said determining motion information of the binocular camera from moment t−1 to the moment t comprises:
determining the motion information of the binocular camera according to the location information of other objects except for the obstacle to be detected in the first view and the location information of the other objects in the second view.
5. An electronic device, wherein the electronic device comprises a memory and a processor, the memory is coupled in the processor and is configured to store computer software codes, and the computer software codes are configured to control the processor to execute the method according to claim 1 .
6. A non-transitory computer storage medium, for storing computer software instructions used by the apparatus for detection of false alarm obstacle and containing program codes designed to execute the method according to claim 1 .