Material Classification of Recyclable Containers Using 60 GHz Radar

Abstract

Rather than sending used containers and materials to the landfill, recycling can help lower the human impact on the environment. However, manually sorting the mixture of incoming material can be both costly and potentially harmful to the person carrying out the task. In many cases, manual sorting could be replaced with automation, where a container is sorted by a machine, based on a classification of the container's material. In this paper, we propose a classification algorithm, using radar data, acquired with Acconeer's A121 60 GHz pulse coherent radar, for classifying liquid containers into one of the four classes metal, glass, plastic, or paper. The solution offers a cost-effective system with robust performance, able to predict the type of container with 98% accuracy.

Tasks

Reproductions