Proposal of a Method of Evaluating Positional Accuracy Based on the Modification of the Simple Buffer Method

Abstract

Classically, the quality control based on the positional accuracy of a cartographic document is performed by the statistical comparison between the horizontal and/or vertical coordinates of well identifiable points on the chart, with the respective coordinates of homologous points obtained on a more reliable product or by method of field survey more accurate than that of the analyzed chart. An alternative to using point features is by using linear features, such as line segments or curves. Some advantages of linear features methods compared to the traditional point method are the large amount and good spatial distribution that linear features generally have in cartographic bases and have more geometric information than points. This work aimed to propose a methodology based on the modification of the Simple Buffer method, to enable it possible to obtain values of planimetric positional discrepancies, in metric units, between the homologous test and reference lines. Using Decree No. 89.817 and ET-CQDG as standard, the results obtained with the proposed methodology were compared with the Simple Buffer, Double Buffer and Vertex Influence methods. In the evaluation of the planimetric positional accuracy of an Ikonos orthoimaging using the 1: 10,000 scale, the results obtained showed that the Modified Simple Buffer method was the most restrictive and rigorous method, being advantageous to obtain a metric discrepancy value, allowing to perform the detection of outliers and tendency a posteriori.