%0 Conference Proceedings
%@isbn 978-85-7669-271-3
%T Probabilistic and dictionary-based relaxation techniques applied to a statistical method of edge detection
%D 1993
%A Alves, André Hiroshi Hayashi,
%A Mascarenhas, Nelson Delfino d’Ávila,
%@affiliation Divisão de Processamento de Imagens (DPI) do Instituto Nacional de Pesquisas Espaciais (INPE)
%@affiliation Divisão de Processamento de Imagens (DPI) do Instituto Nacional de Pesquisas Espaciais (INPE)
%E Figueiredo, Luiz Henrique de,
%E Gomes, Jonas de Miranda,
%B Simpósio Brasileiro de Computação Gráfica e Processamento de Imagens, 6 (SIBGRAPI)
%C Recife
%8 19 - 22 out. 1993
%I Sociedade Brasileira de Computação
%J Porto Alegre
%V 1
%P 81-87
%S Anais
%K relaxation techniques probabilistic, statistical method of edge detection, Image Processing.
%X Two relaxation schemes, a probabilistic and a dictionary-based one, applied to edge detection, are described. The problem of edge detection is defined using a statistical approach. The solution, in terms of statistical decision theory, leads to a test among hypotheses of configurations of sets of four pixels (quadruplets). The relaxation schemes are also developed using the quadriplets as labelling objects. The initial probabilities for the label set of each object are synthesized from the values obtained in the statistical tests. The interaction neighborhood adopted for the two methods is the 4-neighborhood. The iterative label probability updating is performed using a classical heuristic procedure in the two schemes. Tests using noisy synthetic and real images are presented. An experimental analysis of convergence to a consistent and non-ambiguous labelling and speed of convergence is performed for the two schemes and results are compared. A change in the dictionary according to a modification in the definition of consistency is proposed and the resulting scheme is tested and compared with the two other ones.
%9 Processamento de Imagens I
%@language en
%3 10 Probabilistic and dicionary.pdf