抄録
This paper aims to measure the positional accuracy of digitized point data by mouse operation. Fifteen operators digitized 225 representative points of buildings in the experiment designed with Latin square. The frequency distributions of X and Y deviations between digitized point locations and the centers of gravity of buildings on the original drawing are more peaked relative to the normal distribution. An analysis of variance and multiple comparison show that the differences among building coverage and among operators are significant. Substituting estimated deviations into a formula made by Teraki(1999), this paper estimates the probability of identifying buildings by the point-in-polygon interpolation.
