Abstract

Morphological transformations like erosion and dilation can be performed optically by convolving the binary input image with a binary kernel and then passing the result through an ideal thresholder/hardlimiter. Generalizations result from using gray-scale convolution kernels. In particular, gray-scale kernels can be used to perform asymmetrical median (AM) filtering on 1-D continuous, binary signals. AM filters are similar to the median filter, except one AM filter removes only positive-going noise spikes and the other filter removes only negative-going noise spikes from binary images. The two AM filter operations are equivalent to the opening and closing operations for a broad class of 1-D signals, but they can be implemented with just one convolution operation instead of a two-step cascade of erosion and dilation operations.

© 1989 Optical Society of America