Efficient model-based recognition of objects using a calibrated image system
A model-based object recognition system operates to recognize an object on a predetermined world surface within a world space. An image of the object is acquired. This image is a distorted projection of the world space. The acquired image is processed to locate one or more local features of the image, with respect to an image coordinate system of the image. These local features are mapped a world coordinate system of the world surface, and matched to a model defined in the world coordinate system. Annotations can be arranged as desired relative to the object in the world coordinate system, and then inverse-mapped into the image coordinate system for display on a monitor in conjunction with the acquired image. Because models are defined in world coordinates, and pattern matching is also performed in world coordinates, one model definition can be used by multiple independent object recognition systems.
1. A method of annotating an image of an object located in a predetermined world space of an image processing system adapted to estimate at least a location of the object in the predetermined world space, the image processing system including a monitor for displaying the image, the method comprising the steps of:
calibrating the image processing system for the predetermined world space by determining a transformation between coordinates in an image coordinate system of the image and coordinates in a world coordinate system of the predetermined world space;
defining an annotation of the object relative to the estimated location of the object in the predetermined world space returned by the image processing system, the annotation being defined in the world coordinate system of the world space;
mapping the annotation into the image coordinate system using an inverse of the transformation between the image coordinate system and the world coordinate system, wherein mapping the annotation comprises piece wise mapping of the annotation by:
segregating the annotation into a plurality of local elements;
mapping each of the local elements into the image coordinate system; and
desegregating the mapped local elements within the image coordinate system; and
displaying the annotation of the object on the monitor in conjunction with the image of the object.
2. A method as claimed in claim 1 , wherein the step of defining an annotation of the object comprises any one or more of:
defining an envelope in the world coordinate system encompassing the estimated location of the object in the predetermined world space;
defining an indicator positioned at the estimated location of the object in the predetermined world space;
defining text information at a predetermined position relative to the estimated location of the object in the predetermined world space; and
defining a wire-frame drawing tracing at least a portion of an outline of the object in the predetermined world space.
3. A method as claimed in claim 2 , wherein defining an envelope comprises defining either one of a polygon and an ellipsoid surrounding the estimated location of the object in the predetermined world space.
4. A method as claimed in claim 2 , wherein the image processing system is further adapted to estimate an orientation of the object in the predetermined world space, and wherein defining an envelope further comprises defining an orientation of the envelope to approximate the estimated orientation of the object in the predetermined world space.
5. A method as claimed in claim 1 , wherein displaying the annotation comprises either one of:
mapping the annotation into an overlay buffer associated with the monitor; and
mapping the annotation directly into a display buffer of the monitor.