On Image Analysis by the Methods of Moments
IEEE Transactions on Pattern Analysis and Machine Intelligence
Index-based object recognition in pictorial data management
Computer Vision, Graphics, and Image Processing
A pictorial index mechanism for model-based matching
Data & Knowledge Engineering
Object recognition based on moment (or algebraic) invariants
Geometric invariance in computer vision
Shape representation and recognition from multiscale curvature
Computer Vision and Image Understanding
Invariance signatures: characterizing contours by their departures from invariance
Computer Vision and Image Understanding
A Spine X-Ray Image Retrieval System Using Partial Shape Matching
IEEE Transactions on Information Technology in Biomedicine
Proceedings of the 27th international ACM conference on International conference on supercomputing
Hi-index | 0.00 |
Shapes in images provide important features for pattern recognition, pattern matching, content-based image retrieval, etc. In this paper, a contour based structural approach is proposed to represent any arbitrary shape in an image by an alpha-numeric string. The proposed method performs decomposition of the region into disjoint sub-regions by detecting the features of the contour such as bend points, vertical & horizontal straight lines, etc. The representation of those sub-regions is achieved by using strings in a six letter alphabet. The representation is invariant to size since it includes a size-based ratio. The drawbacks of existing contour-based shape representation techniques are discussed, and it is shown how this new method overcomes them.