Robust search for image features across image sequences exhibiting non-uniform changes in brightness
First Claim
1. Apparatus for finding a model image within a test image, the apparatus comprising:
- model image means for acquiring a model image;
means, connected to said model image means, for generating a multi-resolution representation of said model image, said multi-resolution representation of said model image including a coarsest single-resolution model image, and at least one single-resolution model image of finer resolution;
means, connected to said means for generating a multi-resolution representation of said model image, for performing edge detection upon at least said coarsest single-resolution model image to provide at least a coarsest model image edge map;
test image digitizer means for acquiring a test image;
means, connected to said test image digitizer means, for generating a multi-resolution representation of said test image, said multi-resolution representation of said test image including a coarsest single-resolution test image, and at least one single-resolution test image of finer resolution;
means, connected to said means for generating a multi-resolution representation of said test image, for performing edge detection upon at least said coarsest single-resolution test image to provide at least a coarsest test image edge map;
means, connected to said means for performing edge detection upon said coarsest single-resolution model image, and connected to said means for performing edge detection upon said coarsest single-resolution test image, for computing a match-metric value for each one of a plurality of relative displacements of the coarsest model image edge map, with respect to said coarsest test image edge map, to provide a match-metric surface in a match space;
means, connected to said means for computing a match-metric value for each one of a plurality of relative displacements, for determining characteristic information and height information for each local maximum of said match-metric surface;
means, connected to said means for determining characteristic information and height information, for selecting acceptable local maxima of said match-metric surface using both said characteristic information and said height information for each local maximum; and
output means, connected to said means for selecting acceptable local maxima of said match-metric surface, for providing position information of said acceptable local maxima to a manipulation device.
1 Assignment
0 Petitions
Accused Products
Abstract
An apparatus and method for two-dimensional search for a model image within atest image that exploits edge-based matching of edge maps, followed by area-based matching of at least coarse resolution edge maps of respective multi-resolution edge-maps, to efficiently and accurately provide the position of at least one instance of the model image. The invention can find a location of the model image within each test image of a sequence of test images, even when there are non-uniform brightness changes in the test image due to process steps or lighting changes. The postion of each verified instance of the model image within the test image can be provided to an alignment device, a measurement device, or any other device that can use such position information.
-
Citations
26 Claims
-
1. Apparatus for finding a model image within a test image, the apparatus comprising:
-
model image means for acquiring a model image; means, connected to said model image means, for generating a multi-resolution representation of said model image, said multi-resolution representation of said model image including a coarsest single-resolution model image, and at least one single-resolution model image of finer resolution; means, connected to said means for generating a multi-resolution representation of said model image, for performing edge detection upon at least said coarsest single-resolution model image to provide at least a coarsest model image edge map; test image digitizer means for acquiring a test image; means, connected to said test image digitizer means, for generating a multi-resolution representation of said test image, said multi-resolution representation of said test image including a coarsest single-resolution test image, and at least one single-resolution test image of finer resolution; means, connected to said means for generating a multi-resolution representation of said test image, for performing edge detection upon at least said coarsest single-resolution test image to provide at least a coarsest test image edge map; means, connected to said means for performing edge detection upon said coarsest single-resolution model image, and connected to said means for performing edge detection upon said coarsest single-resolution test image, for computing a match-metric value for each one of a plurality of relative displacements of the coarsest model image edge map, with respect to said coarsest test image edge map, to provide a match-metric surface in a match space; means, connected to said means for computing a match-metric value for each one of a plurality of relative displacements, for determining characteristic information and height information for each local maximum of said match-metric surface; means, connected to said means for determining characteristic information and height information, for selecting acceptable local maxima of said match-metric surface using both said characteristic information and said height information for each local maximum; and output means, connected to said means for selecting acceptable local maxima of said match-metric surface, for providing position information of said acceptable local maxima to a manipulation device.
-
-
2. A method for finding a model image within a test image, the method comprising the steps of:
-
acquiring a model image; generating a multi-resolution representation of said model image, said multi-resolution representation of said model image including a coarsest single-resolution model image, and at least one single-resolution model image of finer resolution; performing edge detection upon at least said coarsest single-resolution model image to provide at least a coarsest model image edge map; acquiring a test image; generating a multi-resolution representation of said test image, said multi-resolution representation of said test image including a coarsest single-resolution test image, and at least one single-resolution test image of finer resolution; performing edge detection upon at least said coarsest single-resolution test image to provide at least a coarsest test image edge map; computing a match-metric value for a plurality of relative displacements of the coarsest model image edge map, with respect to said coarsest test image edge map, to provide a match-metric surface in a match space; determining characteristic information and height information for each local maximum of said match-metric surface; selecting an acceptable local maximum of said match-metric surface using said characteristic information and said height information for each local maximum; and providing position information of said acceptable local maximum to indicate position of said model image. - View Dependent Claims (3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26)
-
Specification