Towards Unsupervised Weed Scouting for Agricultural Robotics

TL;DR

This paper proposes an unsupervised clustering approach for weed scouting in agriculture, enabling plant identification without prior species knowledge, demonstrated on field data from an agricultural robot.

Contribution

It introduces a novel clustering method using deep learning features that can be deployed in any field without retraining for specific weed species.

Findings

Successfully clustered cotton plants from grasses without prior training

Deep convolutional neural networks improved plant representation

View-tieing enhanced clustering accuracy

Abstract

Weed scouting is an important part of modern integrated weed management but can be time consuming and sparse when performed manually. Automated weed scouting and weed destruction has typically been performed using classification systems able to classify a set group of species known a priori. This greatly limits deployability as classification systems must be retrained for any field with a different set of weed species present within them. In order to overcome this limitation, this paper works towards developing a clustering approach to weed scouting which can be utilized in any field without the need for prior species knowledge. We demonstrate our system using challenging data collected in the field from an agricultural robotics platform. We show that considerable improvements can be made by (i) learning low-dimensional (bottleneck) features using a deep convolutional neural network to represent plants in general and (ii) tying views of the same area (plant) together. Deploying this algorithm on in-field data collected by AgBotII, we are able to successfully cluster cotton plants from grasses without prior knowledge or training for the specific plants in the field.