Abstract

This paper compares two manual measurement
techniques for measuring retinal vessel segment widths: the
kick-points technique and the edge marking technique. An image
set of 164 clear, high-resolution segments was used. The
kick-points approach uses kick points marked by observers
along interpolated cross-sectional intensity profile graphs; the
edge marking method allows observers to nominate the edges on
a zoomed-up image, and interpolates edge positions. The edgemarking
method provides more precise measurements than the
kick-points method, but these are subject to more inter-observer
variability; we speculate that this result is due to differing observer
perceptions of the edge location.