Method and system for determining a spatial transformation employing partial dimension iterative closest point
AssigneeGM Global Technology Operations LLC
InventorsOren Longman, Noam Zac
Life of this patent
Estimating the term…
Abstract
A spatial monitoring system employs a partial dimension iterative closest point analysis to provide improved accuracy for point cloud registration. The partial dimension iterative closest point analysis improves registration accuracy by performing optimization in accordance with an error magnitude of each dimension, wherein dimensions having large initial errors are significantly improved, and dimensions having high initial accuracy are further improved. The registration separately optimizes each dimension using surfaces with contributing information for the optimized dimension.
Claims19
A system for determining an image spatial transformation/image point cloud registration, comprising:
- a spatial sensor and a controller;
- the spatial sensor arranged to capture a plurality of point clouds of a field of view;
- the controller in communication with the spatial sensor, the controller having an instruction set that is executable to:
- identify, via the spatial sensor, a plurality of source points associated with a first point cloud;
- identify, via the spatial sensor, a plurality of target points associated with a second point cloud;
- execute a normal calculation for each of the plurality of target points;
- categorize the plurality of target points into one of an x, y, or z dimension based upon the normal calculation;
- determine an initial transform between the plurality of source points and the plurality of target points;
- for each of the x, y, and z dimensions:
- determine, for the plurality of source points, an association between the plurality of source points and the plurality of target points,
- determine a correlation cost between the plurality of source points and the plurality of target points,
- determine a new transform based upon the x, y, and z dimensions and the correlation cost between the plurality of source points and the plurality of target points, and
- update source points for the x, y, and z dimensions based upon the new transform;
- and
- determine a spatial transformation between the plurality of source points and the plurality of target points based upon updated source points for the x, y, and z dimensions.
A method for determining an image spatial transformation, the method comprising:
- capturing, via a spatial sensor, a plurality of point clouds of a field of view;
- identifying a plurality of source points associated with a first point cloud;
- identifying a plurality of target points associated with a second point cloud;
- executing, via a controller, a normal calculation for each of the plurality of target points;
- categorizing the plurality of target points into one of an x, y, or z dimension based upon the normal calculation;
- determining an initial transform between the plurality of source points and the plurality of target points;
- and
- for each of the x, y, and z dimensions:
- determining, for the plurality of source points, an association between the plurality of source points and the plurality of target points,
- determining a correlation cost between the plurality of source points and the plurality of target points,
- determining a new transform based upon the x, y, and z dimensions and the correlation cost between the plurality of source points and the plurality of target points, and
- updating source points for the x, y, and z dimensions based upon the new transform;
- and
- determining a spatial transformation between the plurality of source points and the plurality of target points based upon updated source points for the x, y, and z dimensions.
Dependencies are inferred from the claim text.