Fusion of obstacle detection using radar and camera
First Claim
1. A vehicle obstacle detection system comprising:
- an imaging system for capturing objects in a field of view;
a radar device for sensing objects in a substantially same field of view, the substantially same field of view being partitioned into an occupancy grid having a plurality of observation cells, the occupancy grid partitioned angularly and radially to form the plurality of observation cells;
a fusion module for receiving radar data from the radar device and imaging data from the imaging system, the fusion module projecting the occupancy grid and associated radar data onto the captured image, the fusion module extracting features from each corresponding cell using sensor data from the radar device and imaging data from the imaging systema primary classifier for determining whether an extracted feature extracted from a respective observation cell is an obstacle; and
at least one secondary classifier for classifying obstacles exterior of the vehicle, wherein weighting is applied to an output of the primary classifier and outputs of the at least one secondary classifier, wherein the weighted output of the primary classifier and the weighted output of the at least one secondary classifier are used to cooperatively identify obstacles in the field of view.
3 Assignments
0 Petitions
Accused Products
Abstract
A vehicle obstacle detection system includes an imaging system for capturing objects in a field of view and a radar device for sensing objects in a substantially same field of view. The substantially same field of view is partitioned into an occupancy grid having a plurality of observation cells. A fusion module receives radar data from the radar device and imaging data from the imaging system. The fusion module projects the occupancy grid and associated radar data onto the captured image. The fusion module extracts features from each corresponding cell using sensor data from the radar device and imaging data from the imaging system. A primary classifier determines whether an extracted feature extracted from a respective observation cell is an obstacle.
20 Citations
21 Claims
-
1. A vehicle obstacle detection system comprising:
-
an imaging system for capturing objects in a field of view; a radar device for sensing objects in a substantially same field of view, the substantially same field of view being partitioned into an occupancy grid having a plurality of observation cells, the occupancy grid partitioned angularly and radially to form the plurality of observation cells; a fusion module for receiving radar data from the radar device and imaging data from the imaging system, the fusion module projecting the occupancy grid and associated radar data onto the captured image, the fusion module extracting features from each corresponding cell using sensor data from the radar device and imaging data from the imaging system a primary classifier for determining whether an extracted feature extracted from a respective observation cell is an obstacle; and at least one secondary classifier for classifying obstacles exterior of the vehicle, wherein weighting is applied to an output of the primary classifier and outputs of the at least one secondary classifier, wherein the weighted output of the primary classifier and the weighted output of the at least one secondary classifier are used to cooperatively identify obstacles in the field of view. - View Dependent Claims (2, 3, 4, 5, 6)
-
-
7. A method for determining an obstacle exterior of a vehicle comprising the steps of:
-
capturing objects in a field of view by an imaging system; sensing objects in a substantially same field of view a radar device, partitioning the substantially same field of view sensed by the radar device into an occupancy grid having a plurality of observation cells, the occupancy grid partitioned angularly and radially to form the plurality of observation cells; receiving, by a fusion module, radar data from the radar device and imaging data from the imaging system, the fusion module projecting the occupancy grid and associated radar data onto the captured image, the fusion module extracts features from each corresponding cell using sensor data from the radar device and imaging data from the imaging system for identifying potential obstacles; and classifying the extracted feature by a primary classifier for determining whether the extracted from a respective observation cell is an obstacle, wherein the primary classifier determines a posterior probability estimation for each respective cell for identifying whether a feature located within each respective cell is an obstacle, and wherein the posterior probability estimation is calculated based on radar and imaging system data. - View Dependent Claims (8, 9, 10, 11)
-
-
12. A vehicle obstacle detection system comprising:
-
an imaging system for capturing objects in a field of view; a radar device for sensing objects in a substantially same field of view, the substantially same field of view being partitioned into an occupancy grid having a plurality of observation cells, the occupancy grid partitioned angularly and radially to form the plurality of observation cells; a fusion module for receiving radar data from the radar device and imaging data from the imaging system, the fusion module projecting the occupancy grid and associated radar data onto the captured image, the fusion module extracting features from each corresponding cell using sensor data from the radar device and imaging data from the imaging system; and a primary classifier for determining whether an extracted feature extracted from a respective observation cell is an obstacle, wherein a posterior probability estimation is determined by the classifier for each respective cell for identifying whether a feature located within each respective cell is an obstacle, and wherein the posterior probability estimation is calculated based on radar and imaging system data. - View Dependent Claims (13, 14, 15, 16, 17, 18)
-
-
19. A vehicle obstacle detection system comprising:
-
an imaging system for capturing objects in a field of view; a radar device for sensing objects in a substantially same field of view, the substantially same field of view being partitioned into an occupancy grid having a plurality of observation cells, the occupancy grid partitioned angularly and radially to form the plurality of observation cells; a fusion module for receiving radar data from the radar device and imaging data from the imaging system, the fusion module projecting the occupancy grid and associated radar data onto the captured image, the fusion module extracting features from each corresponding cell using sensor data from the radar device and imaging data from the imaging system; a preprocessing module for processing data obtained from the radar device, wherein the preprocessing module applies a constant false alarm rate technique to detect a target in the field-of-view of radar, and wherein an output from the preprocessing module is a signal-to-noise ratio of each cell that is provided to the fusion module; and a primary classifier for determining whether an extracted feature extracted from a respective observation cell is an obstacle.
-
-
20. A method for determining an obstacle exterior of a vehicle comprising the steps of:
-
capturing objects in a field of view by an imaging system; sensing objects in a substantially same field of view a radar device, partitioning the substantially same field of view sensed by the radar device into an occupancy grid having a plurality of observation cells, the occupancy grid partitioned angularly and radially to form the plurality of observation cells; receiving, by a fusion module, radar data from the radar device and imaging data from the imaging system, the fusion module projecting the occupancy grid and associated radar data onto the captured image, the fusion module extracts features from each corresponding cell using sensor data from the radar device and imaging data from the imaging system for identifying potential obstacles; applying a temporal smoothing technique to captured images for refining a posterior probability estimate in determining whether a respective feature is an obstacle, the temporal smoothing identifying motion continuity between captured features of time-based images; and classifying the extracted feature by a primary classifier for determining whether the extracted from a respective observation cell is an obstacle. - View Dependent Claims (21)
-
Specification