Invariants for the recognition of planar contour and gray-scale / Images Invarianten zur Erkennung ebener Kontur- und Graubilder

Abstract

The application of measuring techniques in practice tend to incorporate more and more complex sensors associated with the appropriate signal processing. Typical examples are problems in Visual inspection using a video camera as image sensor and applying pattern recognition methods. The recognition of patterns or objects independent of the viewing angle belongs to one of the fundamental tasks which are not yet sufficiently investigated. The paper gives some results for the construction, the existence and theproperties of invariants for the position-independent recognition of planar contour and gray-scale images. We consider modifications of the position caused by planar and spatial motion of rigid objects (Euclidean motion) with Orthographie projection into the camera plane which define corresponding equivalence classes. Dijferent Performance measures of the invariant features lilce completeness, robustness as well as the matter of computational complexity are discussed.

Algorithms for Hardware-Based Pattern Recognition

“…Wagh and Kanetkar [2] presented a general class of nonlinear translation-invariant transforms, which were called circular transforms (CTs). Burkhardt et al [3] proposed a recursive definition of the class CT, which can be used for a simple mathematical description of the transform. The well-known R(apid) and B(inary) transforms are members of the abovementioned class of transforms.…”

“…The separability properties of nonlinear transforms are generally speaking incomplete. Therefore, it is obvious to use group-theory-based methods to improve the separability properties [3,4,5].…”

“…Separability properties of binary pattern. The amount of separable patterns is processed by Polya's counting theory (cf [3]…”