DeepWeeds: A Multiclass Weed Species Image Dataset for Deep Learning

TL;DR

This paper introduces DeepWeeds, a large annotated image dataset of Australian weed species, and demonstrates high-accuracy deep learning models for weed classification to enable robotic weed control in rangelands.

Contribution

It provides the first extensive public dataset of weed images from Australian rangelands and benchmarks deep learning models for weed classification.

Findings

DeepWeeds dataset contains 17,509 images of 8 weed species.

Inception-v3 and ResNet-50 achieved over 95% accuracy.

ResNet-50 can classify in real-time at 53.4 ms per image.

Abstract

Robotic weed control has seen increased research of late with its potential for boosting productivity in agriculture. Majority of works focus on developing robotics for croplands, ignoring the weed management problems facing rangeland stock farmers. Perhaps the greatest obstacle to widespread uptake of robotic weed control is the robust classification of weed species in their natural environment. The unparalleled successes of deep learning make it an ideal candidate for recognising various weed species in the complex rangeland environment. This work contributes the first large, public, multiclass image dataset of weed species from the Australian rangelands; allowing for the development of robust classification methods to make robotic weed control viable. The DeepWeeds dataset consists of 17,509 labelled images of eight nationally significant weed species native to eight locations across northern Australia. This paper presents a baseline for classification performance on the dataset using the benchmark deep learning models, Inception-v3 and ResNet-50. These models achieved an average classification accuracy of 95.1% and 95.7%, respectively. We also demonstrate real time performance of the ResNet-50 architecture, with an average inference time of 53.4 ms per image. These strong results bode well for future field implementation of robotic weed control methods in the Australian rangelands.