An evaluation of machine learning methods for speed-bump detection on a GoPro dataset

Every day, new applications arise relying on the use of high-resolution road maps in both academic and industrial environments. Autonomous vehicles rely digital to navigate when optical sensors cannot be trusted, such as heavy rainfalls, snowy conditions, fog, other situations. These situations increase risks accidents disable potentials real-time mapping sensors. To tackle those problems, we present a methodology automatically map anomalies road, namely speed bumps this study, using an off-the-shelf camera (GoPro) Machine Learning (ML) algorithms. We acquired data over series differently shaped applied three classification techniques: Naive Bayes, Multi-Layer Perceptron, Random Forest (RF). With 96% accuracy, then RF was able identify GoPro dataset automatically. The results show potential proposed developed surveying produce highly-detailed vertical with fast accurate update rate.