Biologically Inspired Hierarchical Model for Traffic Sign Recognition

Abstract

In this research, a study of the Hierarchical Temporal Memory (HTM) model using the Numenta Implementation “Nupic v1.7” and an investigation of its performance in solving the German traffic sign recognition task using standard benchmark data is conducted. Its performance is compared to that of a state of the art Multi-Column Deep Neural Network (MCDNN) that has been proved to exceed the human recognition accuracy for the same task.