Evaluation of Automatic Prediction of Small Horizontal Curve Attributes of Mountain Roads in GIS Environments

Road curve attributes can be determined by using Geographic Information System (GIS) to used in road vehicle traffic safety and planning studies. This study involves analyzing the GIS-based estimation accuracy length, radius number of small horizontal curves on a two-lane rural forest road. The prediction success was investigated digitized raw generalized/simplified segments. Two different roads were examined, involving 20 test groups two control groups, 22 datasets obtained from surveyed based satellite imagery, GIS estimates, field measurements. Confusion matrix tables also evaluate geometry. F-score, Mathews Correlation Coefficient, Bookmaker Informedness Balanced Accuracy investigate performance groups. Kruskal–Wallis analyze statistical relationships between data. Compared Bezier generalization algorithm, Douglas–Peucker algorithm showed most accurate predictions at tolerances 0.8 m 1 m. results show that tolerance level contributes number, radius, length curves, which vary with value. Thus, this underlined importance calculating generalizations following manual digitization.