Assessing Completeness and Classification Accuracy Errors in Topographic Mapping of Porto Alegre

Abstract

Considering the importance of assessing semantic quality in topographic maps, this paper presents a methodology for the identification and quantification of completeness and classification accuracy errors of a select subset of geographical information represented on the topographic maps of the municipality of Porto Alegre. We performed the identification of errors through visual analysis, using orthophotos as reference data and the topographic maps as test data. We analyzed 16 feature classes from the Transportation System and Urban Structure categories, quantified the errors, and created thematic maps to represent their spatial distribution. In total, we identified 1,653,673 features across these categories and found 2,080 errors, 1,845 omission errors, 91 commission errors, and 144 classification accuracy errors. These represent 0.111%, 0.005%, and 0.009% of the total features, respectively, indicating that the topographic mapping demonstrates adequate semantic quality when evaluated based on the analyzed quality parameters and a maximum error threshold of 2.5%.