Some Useful Definitions

alignment: The process of running the PatMax tool to search a run-time image for matches on a trained pattern

boundary point: A point along a feature boundary. A boundary point has a location and an orientation (normal to the feature boundary and in the direction of positive intensity change)

clutter: Extraneous features in a run-time image that are not part of the trained model

deformation: A change in a pattern that cannot be described using a linear transformation

deformation rate: A measure of the degree to which a found pattern is deformed from a trained pattern

degree of freedom: Part of a transformation that can be characterized by a single numeric value such as angle

feature: A continuous boundary between regions of dissimilar pixel values. A feature is represented by a list of boundary points

generalized degree of freedom: A degree of freedom other than x-translation or y-translation. For example: uniform scale, x-scale, y-scale, or rotation.

granularity: A measure of the minimum detected feature size in an image.

pattern: A trained (internal) geometric description of an object you wish to find in run-time images.

run-time image: Image in which to locate instances of the trained pattern (also called the target image)

score: Numerical measure of the similarity between the pattern in the trained model and the pattern in the run-time image

shape description: A geometric shape, described from a class derived from ccShape, representing the high-contrast boundaries of an object in an image. A shape description represents shape model properties (polarity and weight information) for each boundary it defines. These properties can be implicit (represented by a pure shape) or explicit (represented by a shape model). PatMax can be trained using shape descriptions. Also called a geometric description.

transformation: Mathematical representation of the equations that describe the conversion of points from one coordinate system to another coordinate system.