Abstract:
A method for optimization of the passes’ number is considered, which makes it possible to reduce the image processing time when implementing various operations, for example, logical filtering and/or depth mapping. A feature of this method is the use of two passes in the forward and reverse directions. The presented pseudocodes allow understanding the essence of the proposed passages. Evaluation of the method performance, confirmed by the results of simulation modeling, showed a noticeable decrease in the temporal characteristics of processing an image with a size of 3 $\times$ 3.
Keywords:optimization, number of passes, 3 $\times$ 3 window, image filtering, convolutional filter.