A Robust Neural Network Based Object Recognition System and Its SIMD Implementation

Autor:	Alfredo Petrosino, Giuseppe Salvi
Rok vydání:	1999
Předmět:	Artificial neural network Computer science business.industry 3D single-object recognition ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Cognitive neuroscience of visual object recognition Computer vision SIMD Artificial intelligence Invariant (mathematics) business Massively parallel
Zdroj:	Euro-Par’99 Parallel Processing ISBN: 9783540664437 Euro-Par
DOI:	10.1007/3-540-48311-x_135
Popis:	Recognition of objects is a particularly demanding problem, if one considers that each image must be interpreted in milliseconds (usually 30 or 40 frames/second). In this paper we propose a massively parallel object recognition system, which makes use of the multi polygonal approximation scheme for the extraction of rotation and translation invariant shape features, in connection with artificial neural networks for the parallel classification of the extracted features. The system has been successfully applied for recognizing aircraft shapes in different sizes, orientations, with the addition of noise distortion and occlusion. Timings on the Connection Machine 200 are also reported.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_dedup___::97d536350d096a9b04793aedf0ed2cb8 https://doi.org/10.1007/3-540-48311-x_135 Zobrazit plný text záznamu