Inductive generation of icon trees in foveated multi-resolution recognition

• Published in: 7th International Conference on Image Processing and its Applications pp. 275 - 279
• Main Authors: Brugnot, S, Siebert, J.P, Cowan, C.W
• Format: Conference Proceeding
• Language: English
• Published: London IEE 1999
• Subjects: Applied sciences; Computer vision and image processing techniques; Exact sciences and technology; Image processing; Image recognition; Information, signal and communications theory; Integral transforms in numerical analysis; Signal processing; Spatial and pictorial databases; Telecommunications and information theory; object recognition; scale invariance; trees (mathematics); transforms; visual databases; dynamic rule induction; inductive generation; icon-space; image matching; orthogonal linear shifts; output cortical space; visual search-space; interpolation; foveated multi-resolution recognition; sampled views; icon trees; rotational invariance; tree-structured icon database; input image; retino-cortical transform; multi-resolution search; image resolution; Experimental result; Grey level image; Image processing; Classification; Automatic construction; Database; Mapping; Multiresolution analysis

Iconic approaches to object recognition rely upon the direct comparison between an unknown input image with of characteristic views of images of a target subject contained in an icon database. Hence, this approach to recognition is distinguished by making comparisons in icon-space as opposed to maki...