%0 Conference Proceedings
%@doi 10.1109/SIBGRA.2003.1241014
%@nexthigherunit 8JMKD3MGPEW34M/46QU2CS
%@tertiarytype Full Paper
%J Los Alamitos
%I IEEE Computer Society
%E Program Committee,
%@versiontype publisher
%@holdercode {ibi 8JMKD3MGPEW34M/46T9EHH} {isadg {BR SPINPE} ibi 8JMKD3MGPCW/3DT298S}
%@resumeid 8JMKD3MGP5W/3C9JHH2
%@usergroup administrator
%@usergroup marciana
%3 corresponding geometric.pdf
%B Brazilian Symposium on Computer Graphics and Image Processing, 16 (SIBGRAPI)
%X In this paper we present further results of the application of Dempster-Shafer Theory for uncertainty reasoning in corresponding distorted images in Computer Vision. In previous work Silva and Simoni [12], the model was applied to correspond radiometric distorted images, that is, images presenting differences in brightness and contrast, as an extension of the work developed in Silva and Simoni [11]. The results showed the model is robust when dealing with pairs of non-equalized images and encouraged us to try to correspond geometric distorted images, that is, pairs of images in which one is rotated in relation to the other. In the conducted experiments, the right image was rotated to different angles to simulate the geometric distortions desired. The model corresponded only pair of points in the images and it successfully established the correspondence for the rotated images. As in previous works, the correspondence evidences are based on the contextual and structural features of a point, and they are combined by Dempster´s rule of combination in uncertainty reasoning. A search process maximizes the Belief on the combined evidences.
%T Corresponding Geometric Distorted Images using an Uncertainty Inference Method
%@secondarytype PRE CN
%K COMPUTER SCIENCE, Dempster-Shafer theory, Distorted images, COMPUTAÇÃO APLICADA, Teoria de Dempster-Shafer, Imagens distorcidas.
%8 12-15 Oct. 2003
%@visibility shown
%@group LAC-INPE-MCT-BR
%@secondarykey INPE-12957-PRE/8236
%@copyholder SID/SCD
%2 sid.inpe.br/marciana/2004/01.19.11.28.47
%@affiliation Instituto Nacional de Pesquisas Espaciais (INPE)
%@affiliation Universidade Braz Cubas (UBC)
%@project Visão computacional / Desenvolvimento de métodos e sistemas computacionais completos, com qualidade, corretos, seguros e persistentes
%4 sid.inpe.br/marciana/2004/01.19.11.28
%D 2003
%S Proceedings
%A Silva, Jose Demisio Simões da,
%A Simoni, P. O.,
%C São Carlos, SP, Brazil
%@area COMP