Title :
Foveated Object Recognition Using Corners
Author :
Arnow, Thomas ; Bovik, Alan C.
Author_Institution :
Dept. of Electr. & Comput. Eng., Univ. of Texas at Austin, Austin, TX
Abstract :
We present a gray scale object recognition system that is based on foveated corner finding and that uses elements of Lowe´s SIFT algorithm. The principles behind the algorithm are the use of high-information gray-scale corners as features, and an efficient corner- finding strategy to find them. The system is tested on a set of tool and airplane images and shown to perform well.
Keywords :
edge detection; object recognition; transforms; Lowe SIFT algorithm; foveated corner detection; gray scale object recognition; scale invariant feature transform; Object recognition;
Conference_Titel :
Image Analysis and Interpretation, 2008. SSIAI 2008. IEEE Southwest Symposium on
Conference_Location :
Santa Fe, NM
Print_ISBN :
978-1-4244-2296-8
Electronic_ISBN :
978-1-4244-2297-5
DOI :
10.1109/SSIAI.2008.4512283
