Issue No. 02 - February (1991 vol. 13)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/34.67649
<p>A new parallel algorithm is proposed for fat image labeling using local operators on image pixels. The algorithm can be implemented on an n*n mesh-connected computer such that, for any integer k in the range (1, log (2n)), the algorithm requires Theta (kn/sup 1/k/) bits of local memory per processor and takes Theta (kn) time. Bit-serial processors and communication links can be used without affecting the asymptotic time complexity of the algorithm. The time complexity of the algorithm has very small leading constant factors, which makes it superior to previous mesh computer labeling algorithms for most practical image sizes (e.g. up to 4096*4096 images). Furthermore, the algorithm is based on using stacks that can be realized using very fast shift registers within each processing element.</p>
computerised picture processing; parallel architectures; computational complexity; bit-serial processors; image labeling; local operators; mesh-connected computers; parallel algorithm; communication links; asymptotic time complexity; stacks; very fast shift registers; computational complexity; computerised picture processing; parallel algorithms; parallel architectures
H. Alnuweiri and V. Prasanna Kumar, "Fast Image Labeling Using Local Operators on Mesh-Connected Computers," in IEEE Transactions on Pattern Analysis & Machine Intelligence, vol. 13, no. , pp. 202-207, 1991.