The objective of this paper is to determine whether the metric and thematic values generated from raster-based image objects are significantly different from “true” values provided by existing vector data sets.

For such a purpose, a method of defining boundaries of image objects based on vector data is proposed. The new method uses a vector square grid for pixel representation. Starting at a given pixel size, squares at boundaries are divided into smaller pieces to create realistic image objects boundaries which better reproduce the shape and appearance of real world objects. The proposed approach was applied to extract geometric and biophysical properties of agricultural plots from remotely sensed imagery. Vector-based and raster-based values were compared using paired t-tests which determined the level of statistical significance of the differences.