A data parallel algorithm is presented for spatial data (raster-to-vec
tor) conversion. It operates on pure two-dimensional raster images and
derives a fully topological vector data set, where polygons are defin
ed by homogeneity criteria. The algorithm is implemented in data paral
lel C, and makes extensive use of scan functions that combine communic
ation and computation along grid axes. Code segments listing each step
in building a hierarchy of points, lines, and polygons are listed. Pe
rformance evaluations on a Connection Machine-2 reveal that run-time i
s not sensitive to the number of polygons in the image, and increases
sublinearly with grid size.