A pipelined architecture for image segmentation by adaptive progressive thresholding

A special purpose VLSI architecture for the real-time segmentation of endos copic images is proposed in this paper. The architecture is based on pipeli ned implementation of a new algorithm named adaptive progressive thresholdi ng (APT) that segments the darkest region of an endoscopic image representi ng the gastrointestinal lumen. This segmentation process is an extension of a comprehensive statistical technique based on linear discriminant analysi s for partitioning the image. The APT algorithm is mapped onto a linear pip elined array of simple processing elements with each element of a particula r segment communicating with its neighbours. The APT architecture is partit ioned based on various sequential functions involved in the segmentation pr ocess and these functional modules are organised in a pipelined fashion acc ording to their hardware feasibility. Currently, a prototype of the VLSI ar chitecture for an image size of 256 x 256 is designed and built. The functi onal simulation results obtained in the APT architecture are encouraging. ( C) 1999 Elsevier Science B.V. All rights reserved.