STATISTICAL EVALUATION OF SEQUENTIAL MORPHOLOGICAL OPERATIONS

In order to properly apply sequential morphological operations to rand om signals in applications concerned with noise suppression, we have e xamined their statistical properties using different structuring eleme nts. The performance of hat and triangular structuring elements has be en evaluated for signals with uniform, Gaussian, and Rayleigh density functions. In particular, the statistical properties of sequential mor phological operations (i.e., dilation, closing, dos-erosion, and clos- opening) are examined as a function of the parameters of the structuri ng element through Monte Carlo simulation, which overcomes the statist ical dependency problem arising in the processed signal at different s tages of morphological operations. The simulated results and their sta tistics (mean, variance, and skewness) present an interpretation of th e signal root, biasing effects, and noise suppression capability of mo rphological filters.