Speed Bump Detection, A Time and Feature Selection Analysis

This paper was intended to describe a method for intelligent road monitoring by combining genetic algorithms and logistic regression applied to speed bumps detection. This method was structured by means of a time analysis and selection of features considering six classification techniques. These alternatives were compared with a previous study that analyzed a data set with more than 14,000 road samples. The results presented AUC values of 0.992 and 0.989, suggesting that genetic algorithms and logistic regression could achieve improvements to increase accuracy and minimize false positives in the speed bumps detection.