Road Surface Classification Based on Radar Imaging Using Convolutional Neural Network

The development of a surface recognition system for automobiles is an important and yet unsolved task. The objective of the current study is to demonstrate the benefit of millimeter wave radar for surface discrimination for automotive sensing through the consideration of a novel approach to surface classification based on the analysis of real road surface images acquired using the 79 GHz imaging radar. The proposed experimental method offers good surface categorization accuracy when used with a convolutional neural network. A Convolutional Neural Network (CNN) is a Deep Learning algorithm which can take in an input image, assign importance (learnable weights and biases) to various aspects/objects in the image and be able to differentiate one from the other. The convolution neural networks which learns distinctive features for each class by itself.

Keywords: Millimeter wave radar, radar remote sensing, radar imaging, electromagnetic scattering, convolutional neural network.