Title :
Shape identification using new moment-based descriptors
Author_Institution :
Sch. of Appl. Sci., Nanyang Technol. Inst., Singapore
Abstract :
A new method of using the moment invariants for the position independent identification of 2-D shapes is proposed. The model of a prototype object is described by a family of shapes. The shapes are created by occluding the object by circles (of different radius) located in the object´s centre of gravity. The moment invariants of such shapes are functions (they are called m-invariant functions) of a parameter describing the size of circles. Using m-invariant functions it is possible to create from a single moment invariant many features describing an object, so there is no need to use the moments of higher order (which are sensitive to minor shape deformations). The application of the proposed method to the quality inspection is discussed
Keywords :
computer vision; image recognition; object recognition; 2D shapes; centre of gravity; m-invariant functions; minor shape deformations; moment invariants; moment-based descriptors; occlusion; position independent identification; prototype object; quality inspection; shape identification; two dimensional shapes; Application software; Computer vision; Digital cameras; Gravity; Image analysis; Inspection; Object recognition; Parameter estimation; Prototypes; Shape;
Conference_Titel :
TENCON '94. IEEE Region 10's Ninth Annual International Conference. Theme: Frontiers of Computer Technology. Proceedings of 1994
Print_ISBN :
0-7803-1862-5
DOI :
10.1109/TENCON.1994.369286